Glasgow Theses Service http://theses.gla.ac.uk/
Ross, Susan (2016) The standardisation of Scottish Gaelic orthography 1750-2007: a corpus approach. PhD thesis http://theses.gla.ac.uk/7403/ Copyright and moral rights for this thesis are retained by the author A copy can be downloaded for personal non-commercial research or study, without prior permission or charge This thesis cannot be reproduced or quoted extensively from without first obtaining permission in writing from the Author The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the Author When referring to this work, full bibliographic details including the author, title, awarding institution and date of the thesis must be given.
The Standardisation of Scottish GaelicOrthography 1750-2007: A Corpus Approach
Susan Ross
M.A., M.Litt.
Submitted in fulfillment of the requirements for the Degree of Doctor ofPhilosophy
School of Humanities / Sgoil nan Daonnachdan
College of Arts / Colaiste nan Ealan
University of Glasgow / Oilthigh Ghlaschu
Jan 2016
2
Abstract
This thesis investigates the standardisation of Modern Scottish Gaelic orthography
from the mid-eighteenth century to the twenty-first. It presents the results of the
first corpus-based analysis of Modern Scottish Gaelic orthographic development
combined with an analytic approach that places orthographic choices in their
sociolinguistic context.
The theoretical framework behind the analysis centres on discussion of how the
language ideologies of the phonographic ideal, historicism, autonomy,
vernacularism and the ideology of the standard itself have shaped orthographic
conventions and debates. It argues that current spelling norms reflect an
orthography that is the result of compromise, historical factors and pragmatic
function.
The research uses a digital corpus to examine how three particular features have
been used over time: the dialect variation between <eu> and <ia>; variation in s
+ stop consonant clusters (sd/st, sg/sc, sb/sp); and the use of the grave and acute
accents. Evidence is drawn from the Corpas na Gàidhlig electronic corpus created
at the University of Glasgow: the sub-corpus used in this study includes 117
published texts representing a period of over 250 years from 1750 to 2007, and a
total size of over four and a quarter million words. The results confirm a key period
of reform between 1750 and the early nineteenth century, and thereafter a
settled norm being established in the early nineteenth century. Since then, some
variation has been acceptable although changes and reform of some features have
centred on increasing uniformity and regularisation.
3
Table of Contents
Abstract...................................................................................... 2
List of Tables................................................................................ 7
List of Figures ............................................................................... 8
Acknowledgements......................................................................... 9
Author's Declaration....................................................................... 10
Abbreviations .............................................................................. 11
1 Introduction........................................................................... 12
2 Theories of Writing and Spelling................................................... 14
2.1 The Study of Writing Systems.................................................. 14
2.1.1 Writing and Speech......................................................... 14
2.1.2 Writing is Graphic Communication....................................... 15
2.1.3 The Phonographic Ideal.................................................... 17
2.1.4 Real Writing ................................................................. 17
2.2 The Study of Orthography ...................................................... 18
2.2.1 Standardisation ............................................................. 19
2.2.2 The Sociology of Spelling .................................................. 24
2.3 The Study of Scottish Gaelic Orthography ................................... 27
2.4 Writing and Reversing Language Shift ........................................ 33
2.4.1 Reversing Language Shift (RLS) ........................................... 34
2.4.2 Language Planning for Endangered Languages ......................... 34
2.4.3 Prior Ideological Clarification............................................. 35
2.4.4 How can orthography help RLS? .......................................... 36
2.4.5 How can a standard orthography hinder RLS? .......................... 37
2.5 Studies of Spelling Reforms .................................................... 38
2.5.1 Why Reforms Fail ........................................................... 40
3 Scottish Gaelic........................................................................ 42
3.1 Introduction ...................................................................... 42
3.2 Literacy in Gaelic Scotland..................................................... 45
3.2.1 Ogham........................................................................ 45
3.2.2 The Latin Alphabet and Gaelic Script ................................... 45
3.2.3 Middle Gaelic and Classical Gaelic ....................................... 47
3.2.4 The Seventeenth Century ................................................. 50
3.2.5 The Eighteenth Century ................................................... 51
3.2.6 The Nineteenth century ................................................... 54
3.2.7 The Twentieth Century .................................................... 60
3.2.8 The GOC Era................................................................. 62
4
4 Ideologies in Gaelic Spelling........................................................ 64
4.1 'Standard' Language in the Scottish Gaelic context ........................ 64
4.1.1 A Codified Prestige Standard ............................................. 67
4.1.2 Codified Conventions....................................................... 70
4.2 The Phonographic System ...................................................... 72
4.2.1 Phonography and Vernacularisation ..................................... 72
4.2.2 The Phonographic Ideal in Scottish Gaelic.............................. 73
4.2.3 Phonography as Gaelic Superiority....................................... 75
4.2.4 Phonography, Dialects and Uniformity .................................. 77
4.2.5 The Compromises of Phonography ....................................... 78
4.3 Dialects and Acceptable Variation ............................................ 78
4.3.1 The Curbing of Dialects .................................................... 79
4.3.2 Poetry and Folklore ........................................................ 81
4.3.3 Unacceptable Variation.................................................... 82
4.4 Etymology and Historicism ..................................................... 83
4.4.1 Etymological Principle ..................................................... 83
4.4.2 Etymology and Compromise............................................... 85
4.4.3 Summary ..................................................................... 86
4.5 The Fear of Illegitimacy ........................................................ 87
4.6 Summary and Conclusions ...................................................... 91
5 Methodology .......................................................................... 93
5.1 The Methodology of Corpus Linguistics....................................... 93
5.1.1 What is Corpus Linguistics?................................................ 93
5.1.2 Corpus use in Scottish Gaelic Studies.................................... 94
5.1.3 Corpus Linguistics for Analysing Orthographic Standardisation ...... 95
5.2 Corpus Documentation.......................................................... 96
5.2.1 Corpus Construction........................................................ 96
5.2.2 Corpus Description ......................................................... 98
5.3 Analysis ..........................................................................103
5.3.1 WordSmith Tools...........................................................103
5.4 Summary .........................................................................106
6 Acceptable Variation: <eu> and <ia>.............................................107
6.1 The Breaking of Long /eː/.....................................................107
6.2 <eu> and <ia> in the Literature ..............................................108
6.2.1 The Codification of <eu> .................................................108
6.2.2 Dialects and Acceptable Variation ......................................110
6.2.3 <eu> and <ia> in GOC .....................................................113
6.3 <eu> and <ia> in the Corpus ..................................................114
6.3.1 Lemma Frequency .........................................................114
5
6.3.2 ceud vs ciad ................................................................115
6.3.3 beul vs bial .................................................................120
6.3.4 sgeul vs sgial ...............................................................125
6.3.5 geur vs giar .................................................................130
6.3.6 feur vs fiar .................................................................134
6.4 Summary and Conclusions .....................................................137
7 Uniformity vs Continuity: <st>, <sp> and <sg>..................................140
7.1 Introduction .....................................................................140
7.2 Corpus Analysis of s + stop consonants clusters............................141
7.3 The <sc> and <sg> clusters ....................................................144
7.3.1 Variability in <sc> and <sg> ..............................................144
7.3.2 Initial <sc> and <sg>.......................................................145
7.3.3 Medial and final <sc> and <sg> ..........................................147
7.3.4 Summary of <sc> and <sg> ...............................................151
7.4 The <sb> and <sp> clusters....................................................151
7.4.1 Variability in <sb> and <sp> ..............................................151
7.4.2 Initial <sb> and <sp> ......................................................152
7.4.3 Medial and final <sb> and <sp> ..........................................153
7.4.4 Summary of <sb> and <sp> ...............................................155
7.5 The <st> and <sd> clusters ....................................................155
7.5.1 Variability in <st> and <sd> clusters ....................................155
7.5.2 Corpus Analysis of <st> and <sd> ........................................157
7.5.3 Initial <sd> and <st>.......................................................157
7.5.4 Medial and final <sd> and <st> ..........................................159
7.6 Summary and Conclusions .....................................................169
8 Simplicity vs Phonography: the accent ..........................................171
8.1 A History of the Accent in Scottish Gaelic Orthography ..................171
8.1.1 Position 1: the acute accent only .......................................172
8.1.2 Position 2: the grave accent only .......................................174
8.1.3 Position 3: the grave with the acute for <e> ..........................175
8.1.4 Position 4: the grave with acute on <e> and <o> .....................178
8.1.5 Position 5: no accents.....................................................179
8.1.6 GOC and the Removal of the Acute.....................................180
8.1.7 The Practice of Accents ..................................................182
8.2 Corpus Analysis of Accent Use ................................................183
8.2.1 Accents with <e> ..........................................................183
8.2.2 Accents with <o> ..........................................................188
8.2.3 Accents with <i> ...........................................................193
8.2.4 Accents with <u> ..........................................................197
6
8.2.5 Accents with <a>...........................................................200
8.3 Summary and Conclusion ......................................................205
9 Conclusions ..........................................................................207
9.1 Main Arguments.................................................................207
9.2 Main Findings....................................................................208
9.2.1 Which language ideologies lie behind the choices to argue for reformor continuity in the spelling of Scottish Gaelic? ................................208
9.2.2 What can a digital corpus tell us about how Gaelic has been spelledhistorically? ..........................................................................209
9.3 Further Research ...............................................................210
Appendix 1: Corpus Metadata ..........................................................213
Appendix 2: DASG metadata example.................................................240
Bibliography ...............................................................................244
Corpus Texts............................................................................244
References ..............................................................................251
Online Sites and Software ............................................................265
7
List of Tables
Table 5-1 Metadata features for corpus texts......................................... 97Table 5-2 Corpus data by decade ....................................................... 99Table 5-3 Corpus data by genre ........................................................100Table 5-4 Corpus figures by dialect....................................................101Table 6-1 Frequency of eu/ia word pairs in the corpus ............................114Table 7-1 Examples of is attached to following consonant ........................141Table 7-2 Compounds with medial <sc>, <sg> and <sb> ............................142Table 7-3 Compounds with <sbh> ......................................................142Table 7-4 Words that only appear as <sg-> ...........................................145Table 7-5 <sc-> lemmas and their <sg-> equivalents................................146Table 7-6 <sg> only lemmas in more than 10 texts..................................148Table 7-7 Variation in <sc>/ <sg> forms ..............................................148Table 7-8 <sc> and <sg> lemma with low frequency ................................149Table 7-9 Dates for occurrences of bascaid and basgaid ...........................149Table 7-10 Prefixed as- and eas- before <c>. ........................................150Table 7-11 Frequency of deisciobal....................................................150Table 7-12 Lemmas with only <sp> in the corpus....................................153Table 7-13 Lemmas with medial <sb> in the corpus and their <sp> variants ....154Table 7-14 Corpus occurrences of initial <sd> .......................................158Table 7-15 Instance of sdiuir outwith the corpus....................................158Table 7-16 Lemmas predominantly <sd> ..............................................160Table 7-17 Lemmas predominantly <st> ..............................................161Table 7-18 Loan words from Latin containing <st> or <sd> ........................162Table 7-19 Bèisd and bèist in the corpus .............................................163Table 7-20 English loans appearing as <st> ...........................................164Table 7-21 English loans appearing with both <sd> and <st> ......................165Table 7-22 English loans with /ʤ/.....................................................166Table 7-23 Frequency of an-dràsda/an-dràsta in the corpus ......................167Table 7-24 Lemmas containing <sd> post-1998 ......................................167Table 7-25 <sd> and <st> clusters created by syncope .............................168Table 8-1 No. of accented vowels in Tiomnadh Nuadh (1767).....................176Table 8-2 Five most frequent lemmas with <è> grave ..............................183Table 8-3 Five most frequent lemmas with <é> acute..............................183Table 8-4 Frequency of fein before 1800 .............................................184Table 8-5 Nineteenth-century frequencies of fein ..................................185Table 8-6 Twentieth-century frequencies of fein pre-GOC ........................186Table 8-7 Twentieth-century frequencies of fein post-GOC .......................187Table 8-8 Frequency of fein in texts after 2000 .....................................188Table 8-9 Five most frequent lemmas with <ò> grave ..............................189Table 8-10 Five most frequent lemmas with <ó> acute ............................189Table 8-11 Frequency of mòr and mór from 1751 - 1900...........................190Table 8-12 Frequency of mòr and mór from 1900 - 1980 ...........................192Table 8-13 Frequency of mòr and mór post-GOC....................................193Table 8-14 Five most frequent lemmas with grave <ì>: whole corpus results...194Table 8-15 Five most frequent lemmas with acute <í>: whole corpus results ..194Table 8-16 Five most frequent lemmas with <í> lemmas: without Ais-éiridh(1751) ......................................................................................194Table 8-17 Comparative frequency of <ì> and <í>...................................195Table 8-18 Five most frequent <ì> lemmas...........................................197Table 8-19 Five most frequent <u> lemmas ..........................................197
8
Table 8-20 Comparative frequency of <ù> and <ú> .................................199Table 8-21 Five most frequent lemmas with accented <a> ........................202Table 8-22 Frequency of <á> lemmas with >10 occurrences.......................202Table 8-23 Frequency of a/as in nineteenth-century texts ........................203Table 8-24 Frequency of a/as between 1900 and 1980 .............................204Table 8-25 Frequency of a/as post-GOC ..............................................205
List of Figures
Figure 5-1 WordSmith concordance....................................................104Figure 5-2 Sample of spreadsheet data ...............................................105Figure 5-3 COUNTIFS formula in Excel ................................................106Figure 6-1 Frequency of ceud/ciad in 18th-century corpus texts .................115Figure 6-2 Frequency of ceud/ciad in 19th-century corpus texts .................116Figure 6-3 Frequency of ceud/ciad between 1900 and 1969.......................118Figure 6-4 Frequency of ceud/ciad between 1970 and 1980.......................119Figure 6-5 Frequency of ceud/ciad post-GOC ........................................120Figure 6-6 Frequency of beul/bial in 18th-century corpus texts ..................121Figure 6-7 Frequency of beul/bial in 19th-century corpus texts ..................122Figure 6-8 Frequency of beul/bial between 1900 and 1980 .......................124Figure 6-9 Frequency of beul/bial between 1981 and 2005 .......................125Figure 6-10 Frequency of sgeul/sgial in 18th-century corpus texts...............126Figure 6-11 Frequency of sgeul/sgial in 19th-century corpus texts...............127Figure 6-12 Frequency of sgeul/sgial between 1900 and 1980 ....................129Figure 6-13 Frequency of sgeul/sgial post-GOC .....................................130Figure 6-14 Frequency of geur/giar in 18th-century corpus texts. ...............130Figure 6-15 Frequency of geur/giar in 19th-century corpus texts ................131Figure 6-16 Frequency of geur/giar between 1900 and 1980 ......................132Figure 6-17 Frequency of geur/giar post-GOC .......................................133Figure 6-18 Frequency of feur/fiar 'grass' in 18th-century corpus texts .........134Figure 6-19 Frequency of feur/fiar 'grass' in 19th-century corpus texts .........135Figure 6-20 Frequency of feur/fiar 'grass' in 20th-century texts pre-GOC.......136Figure 6-21 Frequency of feur/fiar, 'grass', post-GOC ..............................137Figure 7-1 Texts using <sd> and <st> 1980-2007 .....................................166
9
Acknowledgements
I am grateful to many people for their support and guidance throughout the period
of this study. I would like to sincerely thank my supervisors, Professor Roibeard Ó
Maolalaigh and Dr Wendy Anderson, for their direction, advice and patience.
I am grateful to the Faclair na Gàidhlig project and DASG/ Corpas na Gàidhlig for
allowing me access to their digitised texts for use in this study. I am indebted to
the College of Arts at the University of Glasgow for awarding me a scholarship. I
am equally grateful to Catherine McCaig's Trust and Urras Brosnachaidh na
Gàidhlig for financial support.
This thesis would not have been realised with the unfailing support of my husband
Colin, my parents and my sister. Special thanks also go Nicola Carty and Anna Beck
for being my writing partners in the final stages.
When I first took a short course at Sabhal Mòr Ostaig in 1996 I had no idea of all
that learning and speaking Gaelic would give me or that it would put me on a path
to scholarship. This thesis owes much to all those who have taught me including,
Dr Michel Byrne, Ailig 'Valtos' Dòmhnallach, Dr Sheila Kidd, Iain S. Mac a' Phearsain,
Murchadh MacLeòid (SMO), Hector MacNeil of UCCB, Professor Ken Nilsen,
Professor Cathair Ó Dochartaigh, and Mark Wringe.
10
Author's Declaration
I declare that this thesis is entirely the result of my own work. I have referenced
sources from the work of others where necessary. This thesis has not been
submitted for any other degree at the University of Glasgow or any other
institution.
11
Abbreviations
DASG - Digital Archive of Scottish Gaelic
GOC - Gaelic Orthographic Conventions
NLS – National Library of Scotland
RLS – Reversing Language Shift
SSPCK – Society in Scotland for Propagating Christian Knowledge
Phonetic, phonemic, orthographic representations (following Sebba 2007: xiii):
Phonetic: [ ]Phonemic: / /Orthographic: < >
All translations in the text are my own.
12
1 Introduction
This thesis is the first corpus-based study of the history and standardisation of
Scottish Gaelic orthography. In Scottish Gaelic studies, the discussions around
spelling, its regulation and 'correct' or 'best' usage have aroused much debate and
scrutiny. These discussions have, however, also been discussions about language
ownership, authority and status. This research began as a project to examine
actual written language; to record changes and patterns in the written system
using the evidence base of an electronic corpus. However, there can be no
understanding of how or why changes and developments have been made without
attempting to analyse the various, often competing, ideologies about language
and writing which have shaped the spelling of Scottish Gaelic. The thesis,
therefore, has a dual aspect: it considers evidence of how Scottish Gaelic has
actually been used, what is termed 'real writing', alongside an awareness of the
sociolinguistics of orthography i.e. the orthographic choices in their social
context.
This study uses the largest corpus available to date for Scottish Gaelic research
with 117 published texts, covering over 250 years from 1750 to 2007, and which
contains over four and a quarter million words. It looks at the modern period from
the mid-eighteenth century, when publishing in Scottish Gaelic began to occur on
a regular basis, to the early twenty-first century.
The main research questions addressed here are:
1. What language ideologies and practices lie behind the choices for reform
or continuity in the spelling of Scottish Gaelic?
2. What can a digital corpus tell us about how Gaelic has actually been spelled
historically?
In attempting to answer these questions, I will consider what relation there has
been between actual practice and ideologies. I aim to contribute to knowledge
about how the processes of standardisation have taken place in Scottish Gaelic
and the ideological background to it.
13
Chapter 2 introduces theoretical frameworks of the study of writing systems,
language ideologies in standardisation, and how the reform and standardisation of
spelling is analysed and studied. Chapter 3 summarises the position of Scottish
Gaelic, how academics have approached the study of its orthography, and reviews
the evidence for literacy as a basis of understanding the cultural and ideological
framework in which spelling has developed. Chapter 4 looks at the existence of
language ideologies in Scottish Gaelic and how they have shaped discussions of the
use of spelling. In Chapter 5 the rationale for the use of an electronic corpus is
described. The thesis then examines three case studies; Chapters 6, 7 and 8
address the varied usage of the digraphs <eu> or <ia>, the contrast between s +
stop consonant clusters, and the use of grave and acute accents on vowels. Finally,
the last chapter brings together the main conclusions of the thesis.
14
2 Theories of Writing and Spelling
In this chapter, I examine previous literature relating to theories of writing
systems, orthography, spelling reforms and the role of literacy and orthography in
reversing language shift. This review outlines the theoretical background to the
present study and allows a contextualisation of the research questions.
2.1 The Study of Writing Systems
This section summarises key points from the study of writing systems. Since the
publication of I.J. Gelb's Study of Writing in 1952, a number of linguists and
semiologists have explored the 'science of writing' as an area of linguistic research.
Theoreticians such as Roy Harris, Florian Coulmas and Gordon Whittaker have,
more recently, sought to establish a theoretical framework for the study of writing
and writing systems. The field aims to go beyond Saussure's position that writing
is purely of interest as a document de langue, positing that it is rather a uniquely
complex form of communication. It also approaches writing beyond the ethno-
centricity of the pre-eminence of the Latin-based alphabet.
The term 'writing system' is ambiguous in both scholarly and lay usage. Whittaker
outlines two interpretations:
1. An independent system of graphic signs (such as the Roman or Latinalphabet) that renders
a. a specific language (e.g., Latin) orb. a group of languages (e.g., Romance)
2. Such a system for a specific language, plus the rules (e.g.,orthographic) that govern its use (Whittaker 2011: 935)
Meaning 1 is often referred to as a script, so for example, Modern Irish and Scottish
Gaelic use the same script, the Latin alphabet, but have different writing systems,
e.g. different orthographic conventions. In this study, 'writing system' will be used
with the second meaning.
2.1.1 Writing and Speech
Within linguistics, it is understood that written language and spoken language are
distinct. In literate cultures, lay users of language often do not appreciate the
15
distinction and this can affect the social context of written language where speech
is corrected on the authority of 'that's how it is written'. Within linguistics by
contrast, speech is often given primacy over writing. In his handbook, Writing
Systems (2003), Coulmas argues that writing systems should not be viewed by
linguists as the 'poor cousin' of speech. He contends that writing develops and
functions as a device separate from speech, but not secondary to it:
Saussure's [...] observation that 'language and writing are two distinctsystems of signs' must always be kept in mind, but the second part ofhis definition, that writing exists for the sole purpose of representingspeech, must be rejected, for writing follows its own logic which is notthat of speech. (Coulmas 2003: 16)
He advises linguists not to underestimate writing or how, in literate cultures,
writing has an impact on the linguistic behaviour of people. Coulmas bases his
theory on four assumptions, which this study will also follow.
• Writing and speech are distinct systems.
• They are related in a variety of complex ways.
• Speech and writing have both shared and distinct functions.
• The bio-mechanics of the production and reception of speech and
writing are different.
2.1.2 Writing is Graphic Communication
If writing is not solely a means of representing speech, what is it? Definitions vary.
Some current theorists, following Roy Harris (1995, 2000), have a broad definition
of what should be included as writing, including musical notation; however, this
has not been widely accepted. A more accepted explanation of writing is that it
is a type of graphic communication; a system of signs with a code that can be
interpreted and related to speech. It is usually visual, but can also be tactile, such
as braille (see Whittaker 2011: 935).
Whittaker identifies three types of graphic communication: writing, notation and
iconography. Notation serves to calculate, order or distinguish units of non-
linguistic data such as music and mathematics where discrete units of meaning,
rather than speech, are represented (Whittaker 2009: 51 and Whittaker 2011:
16
935). Iconography codifies and represents information by means of pictorial
elements, such as in emergency exit signs or emoticons, and has no fixed
relationship to language. Of the three, writing is the most recent of the three in
human development (Whittaker 2009: 52). However, these are not stand-alone
categories: they can influence each other and borrow or incorporate features and
elements from each other. For example, alphabetic writing systems use notational
sets such as numerals. In Gaelic writing, the use of the Tironian sign ┐ for agus,
'and', and '┐c' as etcetera is another example of notation within an alphabetic
writing system.
Florian Coulmas further defines different systems of 'writing' as semiography and
phonography (2003). Semiography assigns to a grapheme (or set of graphemes) a
semantic value of meaning and phonography assigns to a grapheme a
corresponding phoneme or phonetic quality. These are termed morphograms:
signs representing a discrete unit of meaning (a morpheme) or a compound of
such; and phonograms: signs representing a sound or sequence of sounds
(phonemes). Underlying both systems is the core function of writing: to
communicate via visual signs. Just as Whittaker suggests that his different
categories can overlap, Coulmas argues that all writing systems contain both
semiographic and phonographic elements to some degree. Coulmas agrees with
Whittaker that 'Writing systems are, as a rule, mixed systems' (Whittaker 2009:
55). More specifically,
This means that they are composed of at least two of the three signtypes (morphograms, syllabograms, letters), the proportions of whichvary considerably from system to system. (Whittaker 2009: 55)
Current discussion in the field of writing systems is concerned with the
relationship between iconography, notation and writing and its theoretical
implications for the early use and development of writing in ancient civilizations.
For the purpose of this study, however, it suffices to clarify that Scottish Gaelic
is, as all writing systems are, a mixed system based on an alphabetic phonogram
system with elements of notation. This is important as it emphasises that
morphograms are not unknown in phonographic systems. It allows us to understand
that the way we, in literate cultures, interact with text involves interpreting more
than graphic-phonic correspondence; it can also give us grammatical, historical
and other data.
17
2.1.3 The Phonographic Ideal
Coulmas (2003) demonstrates that it is common for both semiography and
phonography to be considered as distinct ideals to which a perfect orthography
should aspire, the relevant one for Scottish Gaelic orthography being the
phonographic ideal. This ideal requires each sound to be represented by one
grapheme and each grapheme to have one phonological realisation. Analysis of
other languages' writing systems and the history of writing shows, however, that
it inevitably remains an ideal rather than an achievable reality and that this is
primarily due to the change and variation inherent in the nature of spoken
language which leads to divergences between graphemes and phonemes.
An illustration of the unrealistic goal of a perfect phonography is given by the
Korean script Han'gŭl. Created in the 15th century by a group of scholars and
detailed in the 1443 document Hunmin Chŏng'ŭm 'Correct Sounds for the
Instruction of the People', Han'gŭl was a purely phonetic script. Unsurprisingly,
however, since its invention many phonetic changes have occurred. Korean
spelling is now more phonemic than phonetic and certain alterations are ignored
in modern spelling. Although it still maintains a close correspondence to the
spoken language, etymological spellings have developed within the writing system
(Coulmas 2003: 156-161).
Despite the impossibility of the phonographic ideal, its influence remains strong
as a way of rating how 'good' a writing system is and it is one of the competing
'principles' or tensions that shape spelling. Where this ideal has explicitly been
argued for in the context of Scottish Gaelic will be documented in Chapter 4.
2.1.4 Real Writing
Coulmas encourages linguists to move beyond the impulse to analyse writing
systems in terms of how much they live up to an ideal.
If [...] we free our grasp on writing from the Western preconceptionthat writing should, really, be a faithful representation of speech, thenthere is little reason to blame writing for whatever discrepancies wediscover in the analysis. (Coulmas 2003: 16)
18
He argues instead for academics to approach their study of writing as it is, not as
it should be.
Writing does not refer exclusively to either thought or sound, and it isquite misleading to consider pure semiography or pure phonography asideals that real writing systems fail to reach. Real writing iscompromise, it is historic, and it is pragmatic. There is no perfect fitbetween the linguistic constructs that are functional in speech andwriting, because writing is static while speech is dynamic. All writingsystems have phonetic and semantic interpretations, they differ in theimportance attached to one or the other. In describing and analysingthe distinctive properties of writing systems and the way in which theyrelate to language these points must be borne in mind. (Coulmas 2003:33) (emphasis mine)
Another argument relevant for the study of the actual use of writing systems is
made by Whittaker who advises that 'we must not confuse the potential of a script
with its purpose, flexibility and actual use at a given time' (2009: 49).
This study aims to consider aspects of the actual use of the Scottish Gaelic writing
systems; to look at how Scottish Gaelic has been spelled along with the expressed
ideals about how it should be spelled, based on an understanding that real writing
is 'compromise, it is historic and it is pragmatic' (Coulmas 2003: 33). While the
written conventions of Scottish Gaelic may have the potential to provide a
consistent phonologically representative system, its actual use will be subject to
social and pragmatic factors and its interpretation by writers and readers. A
writing system is not a closed abstract code but carries the weight of complex
social and historical factors as much as other aspects of human life.
2.2 The Study of Orthography
The term orthography is derived from the Greek ὀρθο (straight, correct) and
γραϕία (that writes, delineates or describes) and this etymology shows how this
term denotes the widely accepted 'correct' or standard way of writing. It is
sometimes used to refer to all 'correct', or socially sanctioned, ways of using the
language including grammar, syntax and pronunciation. This has been rejected in
the Scottish Gaelic context by Donald MacAulay; in his discussion of orthography
he gives two examples that are about 'correct usage':
19
For example the question 'You do not allow people to write Tha e agithe an iasg (instead of Tha e ag ithe an éisg) or Tha e air faicinn amfear sin (instead of Tha e air am fear sin fhaicinn), do you?' has nothingto do with spelling. It has to do with what the questioner considers tobe correct usage in written Gaelic. (1979a: 86)
This focus on 'correctness' demonstrates how closely associated orthography is to
the standardisation of language generally. The study of orthography was for a long
time part of prescriptive linguistic etiquette. Through most of the twentieth
century, the field of linguistics considered it closely related to phonology and
standardised phonemic transcripts. Along with Coulmas' challenge to linguists to
adopt a different scholarly approach to writing systems, recent work has
considered orthography not simply to be a technical transcription matter, but one
with social and cultural relevance. Mark Sebba, in particular, works towards a
theoretical framework for 'a sociolinguistics of orthography' which aims to 'account
for orthographic choices in their social context' (2007: 5). Gaelic orthographic
history has yet to fully account for the sociolinguistic context in which
orthographic choices have been made. As section 3.2 will show, scholarly accounts
of Scottish Gaelic orthography have mainly discussed what spelling should be. This
research aims to begin the work of charting standardisation in Scottish Gaelic
orthography within its sociolinguistic context.
2.2.1 Standardisation
In his work on the history of European vernacular literacy (which he distinguishes
from classical literacy), Fishman (2010) suggests that four 'belief systems' or
'attitudinal dimensions' are at work when a community self-defines their speech
as a language. These are:
• Vitality
• Historicity
• Autonomy
• Standardisation
Vitality is the belief that a language needs a large number of speakers (what that
number actually is, or needs to be, is relative). Historicity is perceived as a marker
of authority and as inherently a 'good thing' and 'is commonly associated with
20
higher prestige and greater visibility' (Fishman 2010: 7-8). Autonomy is required
as a 'real' language must not be considered a dialect of another.
The point of most interest to us here is Fishman's argument that, across Europe,
sociolinguistic attitudes have considered standardisation, or a 'uniformisation' of
a variety across educated members of the community, as a key element in
communities' understanding of what makes their speech a 'language'.
Standardisation interacts with the other factors to form the heart of the language's
legitimacy:
Varieties with the most protected formal or reading/writing functionsand the most firmly instituted governmental or religious functions arealso the same ones that are likely to be regarded as autonomous.(Fishman 2010: 12)
An illustration of the interaction between legitimacy as a language and
standardisation is given by the history of Afrikaans. In the early twentieth century,
promoters of Afrikaans fought for it to be considered a respectable language;
trying to remove its associations with poverty and status as a 'kitchen language' by
reinventing it as a standard language. They succeeded in two generations to turn
'kitchen Dutch' into an Afrikaans of science, technology and higher education
(Alexander 2011).
From a different viewpoint, standardisation arises from the communicative needs
of speakers:
Standardisation of the written language is a process of codifying aparticular set of usages. It arises from a perceived need for uniformitythat is felt by influential portions of society at a given time. A varietyis then selected as a standard, accepted by influential people thendiffused geographically and socially by various means. (Milroy & Milroy1991: 27)
This is not totally at odds with Fishman's framework, as part of the 'perceived need
for uniformity' can be the legitimisation of a language. Alternative motivations,
however, include the need to communicate by written means over distances, for
records to be maintained that can be understood by others in the future and for
mass literacy to be established. Grenoble & Whaley describe one interpretation
of a standard written variety as:
21
[A]n additional variety which should be maximally comprehensible toall speakers. [...] Without standardization, writing becomesidiosyncratic and cannot be interpreted by a large enough body ofspeakers. (Grenoble & Whaley 2006: 130)
In formal matters, such as legal documents, there is a need for cohesiveness and
intelligibility. In the area of formal education, a standard provides a consistent
(and a priori fair) approach, as well as reinforcing the earlier noted aspect of
status and legitimacy:
A written standard often proves to be invaluable in education as well.Teachers need some way of guiding their students in making choiceswhen writing, i.e. some form to teach their students [...] Having astandard orthography can increase the functional domains of alanguage's use, which in turn increases its status within the communityand reinforces community values. (Grenoble & Whaley 2006: 154)
The creation of a written standard variety necessarily entails orthographic
standardisation. Orthography, and discussions around it, therefore also carries the
weight of the belief that standardisation makes a language a 'real' language: the
perceptions of 'correct' spelling contributes to the social legitimacy of the
language as a whole.
In the Scottish Gaelic context, the legitimisation of Gaelic as more than a
'barbarous tongue' was desired by speakers and is expressed from medieval times
(see Chapter 3). Another need was felt by eighteenth century evangelists who
required their fundamental text, the Bible, to communicate to Highland souls, but
in a carefully controlled form that did not stray into varieties that could
misinterpret theological positions. The need for uniformity was felt clearly by
writers, not only at the vernacularisation stage of the mid-eighteenth century,
but as educational needs changed over the centuries. This speaks to the ongoing
nature of standardisation as a process rather than an end point. The variety used
as the basis of the emerging standard was largely that of the influential literati:
the Protestant ministers of the southern districts. The standardised writing system
was diffused by the Bible and the evangelical Gaelic schools.
In this context, threats to Gaelic's orthographic integrity are often felt as threats
to the legitimacy and existence of Gaelic generally. In Chapter 4, I will argue that
22
this fear of illegitimacy is evidenced in the history of Gaelic orthographic
discussions (see Section 4.5).
The Ideology of the Standard
The pragmatic advantages of a standard can often be forgotten, however, in the
perceived inherent superiority of a standard and what is known as the classical
fallacy. Milroy & Milroy (1998) describe how complete standardisation of a living
language is, like the phonographic ideal, unachievable. Instead, they argue that a
standard becomes a variety of the language that real language is measured against
and that standardisation has real social consequences of prejudice.
One of the most apparent is the development of the consciousness and belief that
there are 'right' and 'wrong' forms of language. Roy Harris has described how this
began with ancient societies who advocated the belief that there is only one
proper, correct usage of a language. Harris terms this the 'classical fallacy' (Harris
1980). Milroy describes literate cultures where the classical fallacy is ingrained in
societal beliefs about language as 'standard language cultures'; cultures where
'virtually everyone subscribes to the ideology of the standard language, and one
aspect of this is a firm belief in correctness' (Milroy 2001: 535). This is expressed
in the Classical Gaelic teaching of the bardic schools (known as Irish Grammatical
Tracts) with the terms cóir, 'correct' and lochdach, 'incorrect', prescribing usage
(see Ó Cuív 1965 for description of the linguistic terms in the tracts).
A consequence of this hegemonic correctness of the standard is social
stratification where people have unequal access to a standard form of the
language and where disadvantages based on dialect or educational opportunity
develop (see Milroy & Milroy 1998 for examples of this in anglophone societies).
The social value that writing and orthography has accrued in France has led to a
situation described as a 'Dictatorship of Orthography' (de Closets 2009: 51-2),
where spelling mistakes are considered proof of low education, a lack of personal
diligence and subsequently limit educational success and economic opportunity.
Scottish Gaelic literate culture has not developed in isolation but rather within
the context of bilingualism and therefore alongside the anglophone ideology of
23
the standard. Section 4.1 considers whether the ideology of the standard can be
seen to exist and operate in the same way in Gaelic literate culture.
Variations of the Standard
When Grenoble & Whaley advise on efforts to develop a new standard for an
endangered language, the most important point they emphasise is:
do not choose between features if it is unnecessary, i.e. a writtenstandard can admit a certain degree of variation and still have a highdegree of usability, even for those in the process of learning to readand write. (2006: 133)
There have been recent efforts in the Scottish context to consider orthography in
a way other than as an inevitable history towards a standard, fixed variety. In
Corbett's review of spelling variation in the work of the eighteenth-century poets
Allan Ramsay and Robert Burns, he argues that the variation in Modern Scots
orthography can be considered not as 'chaos' or confusion in relation to a standard
'ideal', but examined by linguists as a complex system in its own right. Corbett
argues that it is possible to see Scots writers' alterations between Standard English
forms and Scots forms, such as sna', snaw and snow, not as conflicting orthographic
systems of English and Scots, but as 'a single system that allows – or even embraces
– variation' for stylistic and prosodic effects (Corbett 2013: 88).
A centralised, uniform standard is not the only language model. An alternative to
standardisation altogether is the concept of a polynomie or a polynomic language.
The concept of a langue polynomique or 'polynomic language' was originally
developed by Jean-Baptiste Marcellesi with regard to his own language, Corsican.
It aims to provide a model for minority languages that is an alternative to the
model of language planning common to majority languages. The standardisation
model from majority languages, it is argued, promotes a prescriptive standard
which reproduces hierarchies and hegemonies within the minority language and
diminishes linguistic diversity. A polynomic approach is a pluralistic model which
rejects a single prestige variety. Although as Julia Sallabank comments,
polynomie as implemented in Corsica does not countenance freevariation, or variation outside certain limits. It seems in effect simplyto support the use of regional variations. (Sallabank 2010: 317)
24
In her research on the Guernesiais community in Guernsey, Sallabank evaluated
the potential success of a polynomic approach there as the community have a
'"folk linguistic" acceptance of regional variation as a source of richness' (2010:
311). In the 2014 Dlùth is Inneach report, which considered the Scottish Gaelic
community's language beliefs, the study participants showed that dialects are also
considered seats of language authority, authenticity and carry sociolinguistic
value in the Gaelic community (see Bell et al. 2014: B129). Also similiarly to
Guernsey, the Dlùth is Inneach research indicates that acceptance of variation in
Scottish Gaelic only applies to where speakers perceive historic or dialectal
legitimacy and there is a lack of tolerance for variation in the spelling of
neologisms and loan words.
In Section 4.3 it is shown that dialectal spelling within standard Scottish Gaelic
orthography has been seen as both detrimental and acceptable as writers argue
over the degree to which 'a written standard can admit a certain degree of
variation' (Grenoble & Whaley 2006: 133). In particular, in 1909 the then Chair of
Celtic at Edinburgh University, Professor Donald MacKinnon, considered dialectal
variation to be acceptable in the literary context of poetry and when illustrating
dialect for linguistic or historical texts (see 6.2.2 for details of MacKinnon's
paradigm). This thesis considers the use of dialectal spelling within a standard
orthography using the case study of <eu> and <ia> spellings in Chapter 6.
2.2.2 The Sociology of Spelling
Signs carry not only linguistic meaning, but also social meaning at thesame time. (Sebba 2007: 7)
As Sebba notes, orthography is a place where different linguistic and cultural
phenomenon intersect; be it social identity, cultural politics, linguistic ideology,
cultural representation, etc. This has also been noted by scholars interested in
the creation of new writing systems,
A key factor for the ultimate success of a written language is itsacceptability to the community, and so social and political issues almostalways enter into the equation. (Grenoble & Whaley 2006: 132)
Written language has been no more immune to the influence of social and political
concerns than any other aspect of human culture over the centuries. Writing, its
25
use and its users, has not existed without the aspects of power and social
hierarchy: who learns the technology of writing, who learns to write, who teaches
what is to be written. As noted by Coulmas and Guerini (2012), the command of
the technology of writing allowed for other cultural hegemonies to establish
themselves, such as sacred religious texts, trade management and empire
administration. Other researchers in writing systems explore the connection
between a written language and its speaker/readers' linguistic identity. The visual
nature and visibility of writing in literate societies means that:
writing systems led themselves easily as symbols of identity. In thissense, there is no symbolically neutral writing system, script ororthography. (Coulmas & Guerini 2012: 437)
This can be simply illustrated by the example of Croatian and Serbian. The spread
of Christianity was tied to the spread of the Roman alphabet and the
predominantly Catholic population of Croatia uses the Roman alphabet. The
spread of Orthodoxy with Cyrillic, however, means Serbian, with a predominant
Orthodox population, uses Cyrillic despite the relationship between the two
varieties being more dialect-like than language-like. These socio-cultural contexts
result in digraphia: two different writing systems for the same language (see
Grenoble & Whaley 2006: 73). In another example. when the Soviet Union decided
to create a written alphabet for the Siberian languages in 1932, they preferred
Roman-based alphabets as Cyrillic was seen as being closely associated with the
previous tsarist, Russian regime. In relation to West African languages, Peter
Unseth has researched the use of different scripts to assert ethnic difference:
By writing their language using a script that is uniquely theirs andinstantly identifiable as distinct from those around them, a number ofethnic communities have created scripts as part of an effort tostrengthen their ethnic identities. (Unseth 2011: 23)
In these countries, this often happened in the period between World War I and
independence, a time of political and social change. This social reality of the
important identity function of a writing system can be often overlooked.
In the early twentieth century, there was a shift from the use of the Latin semi-
uncial script known as clò Gaelach to Roman script for the writing of Irish Gaelic
(Roman type had previously been used in Ireland but its use was not conventional).
26
The organisation An Cuman um Letiriú Shímplí, the Simplified Spelling Society,
was created with the aim of making acquisition of Irish easier in order to facilitate
its survival. The key points proposed were the use of modern Roman type instead
of Irish script and the removal of quiescent letters (called 'dead letters' by Bergin
1911: 13). Osborn Bergin argued against 'national sentiment' by appealing to the
international use of Roman type and the economy and practicality in the
availability of Roman type. When Brian Ó Cuív summarised the position in later
years, he also set up a binary distinction between the 'sentimental' clò Gaelach
script and the 'rational' Roman script:
there developed a prejudice against Roman type which gave rise to themost extraordinary arguments based on sentiment rather than onreason. (Ó Cuív 1969: 25)
A section from Bergin's talk gives a perfect example of orthography (or type at
least) as identity even though Bergin himself does not share the same opinion:
And it is a curious fact that among those who regard the work of ourSociety with suspicion and alarm are several persons who have nointention of learning to read a word of the language. 'We know no Irish,'they complain, 'but at least do not rob us of our alphabet.' (Bergin 1911:20)
This suggests that the potential of a script as an identity marker was so strong in
the Irish context that it could be used in a post-vernacular context. As Doyle
succinctly (2015) summarises:
the reformers failed to take into account the symbolic importance ofIrish having an orthography which was distinct from that of English: thespelling of bhean 'woman' as van was seen as a deplorable Anglicism.(Doyle 2015: 221)
By acknowledging the place of identity in human societies, however, the
'sentimentality' that Ó Cuív dismisses can be seen not as superficial but as a desire,
in common with other peoples, to give themselves a social marker instantly and
clearly distinct from that of other ethnic or cultural groups.
27
2.3 The Study of Scottish Gaelic Orthography
The French linguist Bernard Cerquiglini believes that historians of orthography are
by nature those who wish to change it, although he cautions that they are naive
when they expect their recommendations to be welcomed:
Enclins par métier à simplifier une orthographe dont ils savent si biencomment elle se compliqua, favorables au mouvement de formes quine furent jamais immuables, ces érudits probes, mais naïfs, ne laissentpas de s'étonner chaque fois qu'une de leurs propositions, si modéréesoit-elle, reçoit du public et des écrivains un refus net, suscite descontroverses, déclenche les courroux. (Cerquiglini 2004: 7)
Inclined by their profession to simplify an orthography when they wellknow how it took on its complexity, favourable to the movement offorms which were never immutable, these scrupulous, but naive,scholars never cease to be amazed each time that one of theirpropositions, however moderate it is, receives from the public andwriters a solid refusal, leads to controversies, or provokes anger.
The history of Scottish Gaelic orthography has also generally been approached by
scholars proposing changes and evaluating how best the orthography meets an
ideal, rather than studying it as it is. There are different categories of writings
about the orthography of Scottish Gaelic differentiated by their purpose:
1. simple descriptions (usually called something like 'How toread/pronounce Gaelic');
2. comments on editorial principles to explain an editor's or author'schoices;
3. analyses of individual (often unusual) texts such as the Fernaig MS;4. discussion of the purposes or principles of orthography in order to
propose changes; and5. scholarly descriptions of the history of Scottish Gaelic orthography.
This section will consider scholarly writing covering types 4 and 5. One factor
common to these texts and commentaries is that they are based on the author's
expert, but individual, impressions. In contrast, the use of a digital corpus, as in
the present study, allows us to gather a large data set in order to see how Scottish
Gaelic was actually written in published texts.
An early discussion of Scottish Gaelic orthography is provided by the Rev. Dr
Alexander Stewart (1764–1821), Church of Scotland minister and author of a book
on Gaelic grammar. In the introduction to his influential Elements of Gaelic
28
Grammar (1801, 2nd end. 1812), he details his approach to orthography. Between
the first and second editions there are some small changes, some of them
significant, such as the change in spelling from 'Galic' to 'Gaelic' in the book's title.
The first edition will form the basis of analysis here with significant differences in
the second edition noted where appropriate.
Donald MacKinnon (1839–1914), a native of Colonsay who became the first
Professor of Celtic at the University of Edinburgh in 1882, wrote an article in The
Celtic Review in 1909 in which he offered a general take on the orthographic
system, describing its history and distinctive features. The article is superficially
a brief description of the development of the Scottish Gaelic writing system
beginning with the adoption of the Latin alphabet in preference to Ogham.
MacKinnon describes how the current orthographic system came about by
discussing major points such as the marking of aspiration (i.e. lenition), the
marking of the nasal before consonants and marking broad and slender
consonants. It concludes with a discussion about the use of dialectal forms in
literary works. However, on another level, MacKinnon's purpose in the text is to
maintain the standard by describing its pedigree and expressing disapproval of
non-standard forms. He argues for conservative norms and for the standard as a
set of common, consistent conventions.
Another key text is Donald MacAulay's talk to the Gaelic Society of Inverness in
February of 1977. A native of Lewis, Donald MacAulay (b.1930), was Professor of
Celtic at Glasgow University from 1991 to 1995 and a member of the sub-
committee which was convened in 1976 by the Scottish Education Board and which
produced the Gaelic Orthographic Conventions report of 1981 (SEB 1981).
MacAulay's paper does not aim to give a history of Gaelic spelling, but to discuss
orthography in the context of some theoretical, historical and pragmatic
considerations. It is, in essence, a statement of his position on the need for
established orthographic conventions (he avoids the term 'standard') and an
argument for the need for a sub-committee to codify these. He recommends a
fixed, well-defined norm to aid writing skills. Specifically, he recommends
allowing exceptions to the 'broad to broad' spelling rule to allow for compounds,
participle suffixes and borrowed words; regularising unstressed vowels; using <st>
in place of <sd>; removing the acute accent; and reducing apostrophes.
29
Another Lewis native and Professor of Celtic at Glasgow Univeristy, Derick
Thomson's (1921-2012) brief summary in The Companion to Gaelic Scotland (1994)
is no more than a 'sketch of the topic' (Thomson 1994: 99). However, given
Thomson's influence on orthography as editor of Gairm and in his participation in
GOC 1981 it is useful to consider his account here. Thomson's focus is on the 'strong
thread of continuity' (1994: 99) from Old Irish to modern usage and the change
from inconsistent spellings in Classical Common Gaelic to more standardised use
in the twentieth century. His interest is clearly in the inconsistencies of older
texts, such as Carswell's. Although he mentions alternative orthographies such as
those used in the Book of the Dean of Lismore and the Fernaig Manuscript, the
narrative he presents is of the inevitable and commendable progress towards
consistency. While he does not set out any particular recommendations, he does
illustrate Coulmas' observation in that the aim of his study is to assess how a given
writing system meets an ideal – in this case particularly how it is consistent, fixed
and how exceptions are removed. Thomson's entry in the Companion is the only
one of the scholarly works discussed here which does not argue for one spelling
over another.
The most comprehensive and up-to-date discussion is found in Ronald Black's
'Gaelic Orthography: The Drunk Man's Broad Road' (2010). Black is a Senior Lecturer
in Celtic studies at the University of Edinburgh and Gaelic Editor of The Scotsman.
It is a very thorough biography of the writing system which gives a wealth of
information about the different influential writers, reformers and their texts.
Black's analysis of the Scottish Gaelic writing system is particularly framed in
terms of the author's position vis-à-vis reform.
It is obvious that for Stewart, MacKinnon and MacAulay, the main aim of their
respective texts is to present recommendations for the writing system and its use.
Neither Thomson nor Black's analyses are so explicit in this respect, yet both
express a preference towards uniformity in standardisation that rejects
variability.
In Chapter 2.2, the four belief systems that socially validate a language, according
to Fishman, were presented: vitality (by number of speakers), historicity,
autonomy and standardisation. It is in Stewart's work that they are most explicitly
seen, partly because he spends the most time discussing the nature of Gaelic and
30
of written language and partly because his contemporary social context was so
hostile to Gaelic. He first introduces his work by justifying why Gaelic is a valid
subject for a grammar. He does so by invoking the Vitality belief and arguing that
Gaelic remains the 'common speech of multitudes' (1801: 4). He also covers the
Historicity criterion by emphasising that it is an 'ancient' language (Stewart 1801:
4). MacKinnon also establishes a narrative that gives historical validity to the
Gaelic writing system by detailing its history and development. Stewart also
establishes a firm criterion of Autonomy from Irish Gaelic:
The Scottish writers of Gaelic in general followed the Irish orthography,till after the middle of the last century. However that system may suitthe dialect of Ireland, it certainly is not adapted to the Gaelic of thiscountry. In the Gaelic translation of the New Testament, printed in1767, not only were most of the Irish idioms and inflections, which hadbeen admitted into the Scottish Gaelic writings, rejected, and thelanguage adapted to the dialect of the Scottish Highlands; but theorthography also was adapted to the language. (Stewart 1801: 39)
Stewart, MacKinnon, Thomson, MacAulay and Black all support the existence of a
standard orthography. For Stewart, the Gaelic Bible provides a largely satisfactory
orthographic standard. He frames the desirability of a standard in terms of status;
it is needed in order to 'redeem its credit':
The Gaelic Version of the Sacred Scriptures lately published hasexhibited a model, both of style and orthography, still more agreeableto the purest Scottish idiom, and has a just title to be acknowledged asthe standard in both. Little seems to be now wanting, to confer on theorthography of the Scottish Gaelic such a degree of uniformity, as mayredeem its credit and ensure its stability. (Stewart 1801: 39)
MacKinnon is also quite clear that a standard had existed for Scottish Gaelic since
early in the nineteenth century. The translators of the Gaelic Bible and the
Highland Society Dictionary team were 'capable and scholarly men' (MacKinnon
1909: 11) who reformed the Classical Gaelic norms and by 1828 created,
a fairly uniform standard of orthography, based upon the old, butmodified to suit Scottish requirements. (MacKinnon 1909: 11-12)
It seems that MacKinnon's understanding of 'standard' is of a flexible standard; it
is acceptable even though it is only 'fairly uniform'. MacKinnon's main target, to
which he devotes four pages, is to criticise 'localisms', his word for dialectal or
31
vernacular forms. He criticises the work of the schoolmaster and lexicographer
Neil MacAlpine (1786-1867), the schoolmaster and grammarian James Munro
(c.1794-1870) and the literary editor and translator John MacKenzie (1806–1848)
for being ill-qualified to make pronouncements on orthography. He is particularly
hard on the folklorists John F. Campbell (1821?–1885), Hector MacLean (1818-
1893) and Sheriff Alexander Nicolson (1827–1893) for the way the use of dialectal
forms in their texts (on folk tales, folklore and proverbs) had become so influential
on Scottish Gaelic writing. The danger MacKinnon sees in having too great a role
for dialects within a standard is that outsiders may consider it a loss of linguistic
status. If there is too much variety, the claim to standardisation, one of Fishman's
four pillars of language validity, is undermined, and this in turn undermines the
very status of Gaelic as a valid language. This is discussed further in section 4.3.
An example of this is given by MacKinnon when he warns of localisms disrupting
the maintenance of the standard:
This spread of 'localisms', influenced by Campbell's Popular Tales of theWest Highlands and Nicolson's Gaelic Proverbs, has disrupted themaintenance of the standard and has led to recent eccentricities inGaelic orthography which would have covered us with ridicule if ourlanguage was read by educated foreigners and considered worthy ofnotice. (MacKinnon 1909: 15)
Despite the dangers MacKinnon sees in dialectal representation, the standard he
promotes is a normative standard, not tied to prescriptive rules for the sake of
maintaining social hierarchy, but which maintains continuity with and knowledge
of historical conventions. His article finishes with an exhortation for writers to at
least master the existing standard before reforming and adapting the system.
In Black's history, the value of a fixed orthographic norm is also taken as a given.
Without one, the spelling needs to be 'rescued' (2010: 251). MacAulay's discussion
is the most nuanced with regard to a standard, possibly because he is in the
position of preparing the audience for the reform of the first Gaelic Orthographic
Conventions. MacAulay states clearly that he believes there to be 'no standard
Gaelic in the strict sense of the term' (1976-78: 88), with the strict sense possibly
referring to a prestige dialect for speech as well as writing. He strongly advocates
a normative standard (although he avoids the word 'standard', preferring 'norm' or
'convention'), with room for acceptable variation (which he terms 'deviants').
Unlike MacKinnon, MacAulay does not simply approach standardisation as a matter
32
of status. He discusses the advantage of a standard by its usefulness for written
communication:
If we look at the spelling of different versions and translations of theBible and the shorter catechism we see much variation there, but sincethese texts utilise a basic rough convention the variations by and large,do not make it impossible or even particularly difficult for us to readthe texts. (1976-78: 91)
His other arguments for a 'single agreed norm' are: to prioritise the needs of the
writer as simplicity aids the acquisition of writing skills; it is cheaper (although it
is not clear why this should be the case); it is a normal process of language
development; it is popular: 'Nobody believes spelling should be random' (1976-78:
91-92). He believes that spelling at the time (in the 1970s) was inadequately
following the existing conventions because it was both ill-defined and lacked
authority. He argues for an 'authoritative conventional norm to be established and
agreed upon' for formal orthographic usage within wider literacy awareness of the
appropriateness of informal and dialectal forms to be used 'purposely to mark
written discourse as belonging to a particular area or a particular style' (1976-78:
89). He advocates a strict use of conventional orthographic norms for formal
written discourse and gives the example situations of administration, books on
'serious topics', newspapers, school and university and all formal written prose.
MacAulay hesitantly suggests that this standard orthography could allow for
acceptable variation, as register variation exists in speech:
Our judgements of their correctness should be based not on a singularnorm allowing only one usage but on an assessment of theirappropriateness in the situation in which they appear. [...] We saidabove that orthographic conventions are different in kind from thosegoverning style. However I think we could usefully consider whetherthere is any way in which the notion of appropriateness might applythere also, in spite of the fact that most of the complaints aboutspelling relate to the proliferation of variants. (1976-78: 87)
In both Stewart and MacKinnon's writing the idea of a standard language in the
sense of a prestige variety can be seen. For both writers, this is essential to the
language's status, which in turn is essential to the language's ongoing existence.
For Thomson, MacAulay and Black, there is little evidence of a belief in a prestige
standard variety of Gaelic. Instead, the approach is much more on 'standard' as
33
codified orthographic conventions necessary for promoting written
communication and the practical functioning of the language in the modern world.
MacKinnon's 1909 article can be read as an acknowledgement of Coulmas'
definition of 'real writing' as his history describes Gaelic orthography and defends
it on the basis of it being compromise, historic and pragmatic. MacAulay's analysis
also fits Coulmas' theoretical framework: he specifies that; writing is different to
speech; the main purpose of writing is communication and that writing is graphic
communication. MacAulay expands the latter to mean clear and unambiguous
communication: one of the clear differences between writing and speech for
MacAulay being the importance of clarity in writing as the interlocutor is usually
not present.
It is essential for those marks and their combination to signal theirmessages clearly and unambiguously, for if they do not do so there isno other way in which we can arrive at their meaning: what is writtendown is all we have to go on. (1976-78: 81)
His argument that there should be acceptable variation in spelling is also framed
as pragmatic, as being 'a matter of practicalities' (1976-78: 87).
The scholarly writings raise other questions of language ideology and its relation
to how Scottish Gaelic is, and should be, spelled. This will be discussed in more
detail in Chapter 4.
2.4 Writing and Reversing Language Shift
This section summarises the ways in which orthographic planning and reform
operate within the context of endangered languages and efforts to reverse
language shift. This is the context in which Scottish Gaelic finds itself and the
context in which twentieth-century reforms have taken place. There are various
terms used for languages with a relatively small number of speakers; lesser-used,
non-dominant, minority, minoritised or regional languages. The term endangered
language will be used here as it foregrounds the decline of intergenerational
transmission towards a state where the language is no longer being used as a
vernacular; this is the crucial factor in Gaelic's current position and a key
motivation for revitalisation efforts. The language that replaces the endangered
34
language will be referred to as the dominant language or language of wider
communication (following Grenoble & Whaley 2006).
2.4.1 Reversing Language Shift (RLS)
When a language is lost as a vernacular, the implication is that the population has
shifted its language use to another language. 1 The efforts to revitalise an
endangered language are therefore efforts in reversing language shift (RLS) from
the new (usually economically, socially and politically dominant) language back
to the endangered one. The aim may be to maintain an existing diglossia or to
create a stable bilingual language environment. It is unusual to aim for
monolingualism in an endangered language due to the (perceived) economic and
social limitations it would entail for speakers. Joshua A. Fishman's canonical text
Reversing Language Shift (1991) has guided many theoretical, analytical and
practical responses to language shift. It is not, however, a how-to guide. It is, in
the main, 'about why most efforts to reverse language shift are only indifferently
successful, at best, and outright failures or even contra-indicated and harmful
undertakings, at worst' (1991: 1). Since the publication of this book, the field of
language planning and policy has expanded as scholars and language activists try
to find successful ways to undertake RLS.
2.4.2 Language Planning for Endangered Languages
Language planning is commonly understood as comprising three aspects:
acquisition planning, status planning and corpus planning. It is believed that
increased activity in each sphere can reverse language shift. The first is concerned
with the users of the languages, their numbers and their level of acquisition; the
second with use of the language, increasing social status and prestige and
establishing linguistic rights. Corpus planning is concerned with the workings of
the language itself: which linguistic resources are available to support acquisition
and what form the language will take when status planning efforts require it to be
used in new domains (in terms of vocabulary, register, standardisation, etc.). One
aspect common to many endangered languages is that their use has been
1 Although there is the alternative reality, which thankfully occurs more rarely, that the population ofspeakers themselves have been lost to humanity, e.g. the indigenous Beothuk of Newfoundlandwho were wiped out in the early nineteenth century following European settlement (see Pastore1992).
35
restricted to certain, often considered traditional, domains: the home, immediate
community and areas of non-industrial or pre-capitalist work such as farming,
fishing, hunting and housework. Thus, language planning efforts seek to expand
use of the language into new, or reclaimed, domains: education, professions such
as the law, government and medicine, technological work (engineering, ICT),
broadcasting, etc. Although, as discussed in section 2.2.1.3, the centralised model
of selection, codification, elaboration and dissemination is criticised by
proponents of polynomie such as Jean-Baptiste Marcellesi and Jacques Thiers
(1999) as inappropriate to the needs of an endangered-language community.
In Grenoble and Whaley's guide to language revitalisation Saving Languages (2006)
they review many projects and address orthography as a specific part of
revitalisation efforts. Their focus is specifically on instances where an
orthographic system has been created from scratch. While this is not the case for
Gaelic, the examples they provide are illuminating. For example, in balancing the
priorities for a new writing system, they conclude that 'acceptability stands above
all other priorities in designing a writing system' (2006: 143), as any system cannot
be successful if it does not have acceptance from the community.
2.4.3 Prior Ideological Clarification
Many scholars recommend that before planning for revitalisation begins, it is
critical for community members to have an honest debate in order to understand
and articulate what goals they want to achieve and what are the attitudes and
beliefs they hold which may help or hinder their efforts. This is called 'prior
ideological clarification' (Fishman 1991; Dauenhauer & Dauenhauer 1998) and is
both for the communities' own benefit and for the guidance of any experts or
linguists that become involved. This process is, however, not often attempted or
carried out successfully.2 Fishman also emphasises the importance of popular
linguistic acceptance, or legitimacy, in his book on corpus planning (Fishman
2006), where he cautions language planners to gain a good understanding of the
speech community's dominant language ideology (through prior ideological
2 ‘There has recently been an increase in the publication of edited collections of papers describingcase studies of language revitalization efforts, but these rarely evaluate processes or outcomesor relate these to ideological clarification’ (Austin & Sallabank 2014: 8). See also Fishman 2001:541.
36
clarification) so that they can minimise the risk of engaging in corpus planning
that actively accelerates language shift.
In 2013 a team within the Gaelic research network Soillse, including the author,3
carried out a project named Dlùth is Inneach aimed at carrying out ideological
clarification in order to create a linguistic foundation for future Gaelic corpus
planning efforts (Bell et al. 2014).4 A public consultation involving 39 workshops
and 184 participants gathered information through Focussed Conversations to
establish the dominant language ideology.5
The conventional understanding of corpus planning is that it primarily pertains to
the formal and written forms of the language as they are the most easily
controlled in education, publishing and governmental bodies, although it may also,
however, have an impact on the spoken language. The role of orthography, then,
is generally seen as part of the standardisation effort which aims to increase
literacy, increase the written use of the language so that it can expand to new
domains and consolidate existing ones, and to assist in educational endeavours. It
forms part of literacy efforts in reversing language shift (RLS) which, as noted in
Chapter 3.3, can have a complex role as it offers both prestige to the endangered
language and facilitates acquisition of the dominant language.
2.4.4 How can orthography help RLS?
As discussed in section 2.2.1, the benefits of a standard orthography, or codified
orthographic norms, are the facilitation of written communication across
geographical areas, the facilitation of literacy acquisition, a consistent approach
of teaching and assessment in formal education. There is also the potential for a
shared linguistic identity where various dialects exist. Thus in reversing language
shift, a standard orthography can benefit the community by extending
3 Under the maiden name Bell. For information on Soillse, see www.soillse.ac.uk4 As revitalisation efforts are not only well underway, but institutionalised in Scotland, this was
'ongoing' rather than ‘prior’ ideological clarification.5 The Focussed Conversation is a group facilitation method created by the Institute of Cultural
Affairs which provides a structure for effective communication where everyone in the group hasthe chance to participate. For more on how it was applied as an Ideological Clarificationexercise, see Bell et al. 2014.
37
opportunities for communication as well as the implied status and legitimacy that
contemporary Western societies ascribe to literate cultures.
2.4.5 How can a standard orthography hinder RLS?
Social consequences
One of the most apparent consequences of a standardised writing system is the
development in the minds of speakers that 'standard' equals 'right' or 'correct' and
'non-standard' equals 'wrong' or 'incorrect'. This can hinder RLS efforts as it lowers
the social prestige of oral ability and the confidence of speakers. It may also result
in them withdrawing from intergenerational transmission so as to avoid passing on
'wrong' forms of the language.
In the Scottish Gaelic context, this phenomenon was discussed in Gordon Wells'
study of the community of Uist. It included in its aims to 'find out how Gaelic
speakers (particularly habitually non-literate ones) view Gaelic writing and
whether literacy might be viewed as a pre-requisite in any positive self-evaluation
of competence' (Wells 2011: 8). Although the participants did show a
consciousness of the potential problematic relationships between literate Gaelic
learners and non-literate L1 speakers,6 they attributed these to other people.
Therefore while Wells concludes that
Several participants expressed the concern that a relativeunderconfidence in literacy skills could spill over into an apparentreluctance on the part of fluent speakers to converse in Gaelic, whetherwith Gaelic learners or indeed with other known fluent speakers whomight exhibit signs of “book-learned” language (Wells 2011: 24, boldadded)
The study participants did not testify that they themselves use Gaelic less as a
result of encounters with literate L1 or L2 speakers or give concrete examples of
events they have witnessed. Their concern is based on assumptions of what other
people might think. At this stage, it is not clear whether non-literacy, a standard
orthography, or the GOC reports have affected Gaelic speakers' confidence or
their language use. Further work, with social psychologists, would be needed on
6 Non-literate in Gaelic, but as likely to be literate in English as much as the average British citizen.
38
perceived or self-reported levels of self-confidence regarding language usage in
the Scottish Gaelic-speaking community.
The second social consequence is derived from the prestige of literacy: those who
adhere to and understand the standard are seen as more 'correct' in their use of
language. As people inevitably have unequal access to a standard form, this leads
to social stratification.7 In weighing up these two social consequences, Grenoble
and Whaley conclude that they do not outweigh the benefits because communities
of endangered-language speakers already have social stratification and power
imbalances:
while the introduction of a standard written form will produce newstratification within, and sometimes among, the communities where anendangered language is used, it has the greater potential to rectify themore obvious asymmetry of power that holds between members of thecommunity and those outside it. (Grenoble & Whaley 2006: 156)
2.5 Studies of Spelling Reforms
Coulmas and Guerini classify three kinds of reform (2012: 442):
• Orthography reforms• Script reforms• Writing system reforms
The first, orthographic reform, is the most frequent and least radical, meaning
the adjustment of spelling conventions. A script reform replaces one graphical
code with another. Writing system reforms change the systematic design and the
basic operational units. Coulmas and Guerini suggest that there is one primary
reason that lies behind reform: one which is also true for the reforms of Scottish
Gaelic:
The general motivation for such reforms is to secure the functionalityof the system by simplifying its rules and thus facilitate the task ofchildren becoming literate. (Coulmas & Guerini 2012: 446)
7 See Urla 2012 for an exploration of this process over several years in regard to the ‘Batua’ literaryform of Basque and its impact on speakers.
39
The twentieth-century Gaelic Orthographic Conventions (GOC) had a similar
educational intent. Earlier reforms in the late eighteenth century were interested
in making adults as well as children literate.
Coulmas and Guerini observe that 'Uniformity, transparency and simplicity are
common aims of spelling reforms' (2012: 447). Orthographic reform is often
prompted by the needs of education where uniformity makes assessment fairer
and where teaching is consistent. Transparency can mean adapting spelling to
reflect changes in pronunciation. Simplicity can be a matter of perspective;
however, it is generally believed to aid the acquisition of literacy and reading
skills. Coulmas and Guerini consider these cogent linguistic reasons for reform.
However, they emphasise that there is a lack of clear evidence that design
features help or hinder the acquisition and usability of literacy. They summarise
that 'in regards to literacy, linguistic motivations for writing reform, therefore,
seem to be of secondary importance at best' (Coulmas & Guerini 2012: 448-9).
An example of reform for linguistic reasons is given by France. At the same time
that the Scottish Education Board was concerned about the quality of Gaelic
students' spelling in the 1970s, the French education system was concerned about
the obstacle that a complex orthography put in the way of education. Wider
concerns that education served only an élite not the wider population, meant
spelling had to be considered for change (de Closets 2009: 186-90). In 1988 the
primary teachers' union, SNI-PEGC, published a report on orthography. It focussed
on which errors were causing the most difficulty, on where the writing system was
dysfunctional and caused traps even for those who knew the system well. The
'rectifications' were informed by an understanding that radical change would not
be accepted by the reading public; however, despite efforts to address potential
concerns, a media scandal erupted. The Académie française, which had agreed to
support the reform, overturned its decision. In the end, the recommendations
were not implemented in French schools. However, outside France, the reforms
were welcomed and adopted, for example in Belgium, by the universities and
exam boards. The Office québécoise de la langue française declared itself in
favour in 1991 and in Switzerland a consultation with teachers also resulted in the
adoption of the reforms in 1996. Due to this, the recommendations were gradually
adopted as variants by dictionaries, spell-checkers, until finally, in 2008, they
were accepted in education in France (de Closets 2009: 230-6). The French
40
example demonstrates that conservatism is a strong force, that linguistic authority
can be held by different agents and that the 'common good' aspect means everyone
holds an opinion.
2.5.1 Why Reforms Fail
A reform of spelling is rarely uncontested and just as often can be unsuccessful.
As Whittaker notes,
radical systemic change is comparatively rare in the long history ofwriting and, as a general rule, only take place when a speechcommunity adopts and adapts a writing system. (2011: 938)
He goes on to note that not even the first-century emperor Claudius was able to
introduce additional letters to the Latin alphabet.
The general development of writing is, thus, somewhat reminiscent ofthe punctuated equilibrium model in natural history, according towhich change is not gradual but sudden, followed by lengthy periods ofstability. (Whittaker 2011: 939) (italics original)
That the contested nature of the Scottish Gaelic GOC reforms is not unique is
illustrated through comparison with the modern reforms of Dutch, German and
French orthography in 1995, 1996 and 2000 respectively:
The said reforms do not alter any basic design features of theorthographies in question, pertaining to accent marks, word separation,the use of capital letters, the integration of loan words, as well as someother lesser problems. In all three cases international institutionalefforts were made to realize the reform, yet they met with majorresistance by defenders of the old norm. As a result, rival conventionscoexist. Purists are disturbed by this state of affairs, but many readers,writers and publishers do not seem to be concerned. (Coulmas & Guerini2012: 447)
As Cerquiglini characterises it, orthographic discussions are a battle of ideas:
Pleine de cédilles et de fureur, l'histoire de l'orthographe française estavant tout une bataille des idées. (Cerquiglini 2004: 7)
Full of cedillas and fury, the history of French orthography is above alla battle of ideas.
41
The approach of Cerquiglini to the study of French orthography is to see its history
as an interaction of various language ideologies, linguistic ideals and social
factors. Reforms are contested on these grounds and the ideas behind them can
lose out to the ideas of those of the status quo.
Orthography is often seen as a technical exercise requiring expert linguists, not as
part of language as a social phenomenon. As noted earlier, orthography and
standardisation do not stand apart from social factors. Although Grenoble &
Whaley are mainly concerned with creating orthographies from 'scratch', their
observations on the importance of recognising the social aspects apply also to
orthographic reform:
The importance of sociological factors cannot be overstated. Regardlessof how linguistically and technically sound an orthography might be, itsinitial (and continuing) acceptance by the people for whom it isdesigned is critical in determining its eventual effectiveness and use.Therefore, local leaders and native speakers must be integrally involvedin the process of developing an orthography regardless of theirsupposed linguistic awareness; the creation of a writing system by anoutside linguist or single community member acting independently,without continual local input and feedback, easily leads to a failedorthography. (Grenoble & Whaley 2006: 137-8)
The creation of an orthographic reform by an outside linguist or single community
member 'without continual local input and feedback' has the same chance of
failure.
42
3 Scottish Gaelic
3.1 Introduction
The name 'Gaelic' can be applied to three languages in the Goidelic, or Q-Celtic,
branch of the Celtic languages; Irish Gaelic (Ireland), Scottish Gaelic (Scotland)
and Manx Gaelic (Isle of Man). All three descend from the same parent language,
referred to as Common Gaelic (Jackson 1951). The other branch of the Celtic
language is the Brythonic, or P-Celtic, branch which includes Welsh, Breton,
Cornish and their earlier relatives.
It is generally believed that Gaelic was established in Scotland when the Gaelic-
speaking kingdom of Dál Riada in the north-east of Ireland expanded to include
Argyllshire sometime around 500AD (Thomson 1977: 127).8 Gaelic continued to
expand from that base across Scotland. Ecclesiastical influence followed with the
arrival of Columba and the spread of his Christian monasteries. By the end of the
eleventh century, the Gaels were established over much of what is now within the
borders of modern Scotland; the distribution of Gaelic place-names clearly shows
the presence of Gaelic, at one time or another, excluding only the northern isles
of Orkney and Shetland and the south-east counties of Berwickshire,
Roxburghshire and Selkirkshire (Ó Baoill 2010: 8).9
After this early dominance in the creation of Scotland, Gaelic power and influence
receded at the national level. The royal court stopped being Gaelic-speaking in
the twelfth century and a 'consciously bi-cultural nation' (Gillies 1993: 145)
emerged with Gaeldom established in the Highlands and Islands and the Scots
language and culture in the Lowlands. Various factors contributed to the
subsequent decline of the Gaelic language; the loss of political independence
following the assimilation of clan chiefs and the Lordship of the Isles; the rise in
prestige of Lowland Scots in official and public life; the subsequent rise of the
English language after the royal and political unions with England in 1603 and 1707
8 However, this paradigm has been dismantled by Ewan Campbell as lacking in archaeological,historical or linguistic evidence. He suggests that Argyll was Gaelic-speaking from the Iron Age(Campbell 2001).
9 Recently, this has been challenged with suggestions that Q Celtic speakers may have travelled tomainland Britain from the continent and settled there. See Ó Baoill 2010 (2-3); for a recent up-to-date account of the advent and expansion in Gaelic, see Clancy (2011).
43
respectively; and economic isolation from the core areas of development. Official
and social attitudes became more intolerant of Gaelic and various efforts to
'civilise' and 'improve' people's lives came to imply the eradication of Gaelic.
In the pre-medieval period evidence for Gaelic rests on historical references to
named people and places as there are few surviving Gaelic documents. In the
medieval period there is far more evidence across Scotland and Ireland, yet Gaelic
writing took place in a standardised literary norm across both countries which
reveals little dialectal variation. For this reason, it is difficult to discern the period
of divergence between Irish and Scottish Gaelic. However, it is likely that
divergences have existed from at least the Old Gaelic period (Ó Maolalaigh 2008:
182-97). It was in the seventeenth century and particularly the eighteenth century
that Scottish Gaelic emerged as a distinct written language (see Thomson 1977).
It has been estimated that in the mid-eighteenth century, the start of the period
that this work is concerned with, there were 289,798 Gaelic speakers in Scotland
out of a total population of 1,265,380 (22.9%) (Thomson 1994: 109). However, this
period is also the start of an age of mass migration and forced emigration, known
as the Highland Clearances, which took place from the late-eighteenth century
well into the nineteenth. The effects of this on the population of the Highlands,
and therefore the number of Gaelic speakers, has been extensively studied (see
Hunter 2000 and Withers 1984, 1988) and is ascribed to various socio-economic
factors, agrarian transformation and the growth of capitalism without
industrialisation in the Highlands. In particular, capitalism fundamentally changed
the relationship of Highland leadership to its people from a clan structure to one
of landlords and tenants. The eighteenth and nineteenth centuries also saw
educational and evangelical agencies at work in the Highlands, aiming to improve
the physical and spiritual well-being of the people (for details see Durkacz 1983).
However, when Gaelic language ability was first measured in the census in 1881,
the language was still spoken by nearly a quarter of a million people in Scotland
(Thomson 1994: 111). Social conditions changed again in the 1880s, when the fight
for land rights, known in Gaelic as 'Aimhreit an Fhearainn', the land agitation,
created active political engagement. This led to the Napier Commission and the
Crofting Acts of 1886 and 1892. This activism also created the first Chair of Celtic
in 1882 at the University of Edinburgh and the appointment of Professor Donald
44
MacKinnon. The early twentieth century saw heavy population loss in the First
World War, followed by the continuing trend in economic migration and language
shift to English. Gaelic speakers emigrated to many British colonies and beyond.
Small diaspora communities can still be found in Nova Scotia, Canada.
The establishment of organisations such as An Comunn Gaidhealach provided a
central focus of activity and activism as the twentieth century continued. An
Comunn has run classes, summer camps, published books and school textbooks
and has run the annual national festival, the Mòd, among many other initiatives.
While the number of speakers and the geographical area have both declined, there
has been a rise in the range of activity in education, broadcasting, publishing and
community groups since the 1980s. The results of these may not yet have had their
full effect or, as has also been noted, 'particular initiatives have increased Gaelic
prestige, while the figures of speakers tell their own story' (Thomson 1994: 91).
By 2001, accelerating Gaelic-to-English language shift meant that there were no
monolingual Gaelic speakers left and 90% of the 340,000 people living in the
Highlands and Islands were monolingual English-speakers (GROS 2005).
Today Scottish Gaelic is an endangered language. From the 2011 census in
Scotland, 87,100 people aged 3 and over in Scotland (1.7 per cent of the
population) reported having some Gaelic language skills. The breakdown of
reported language skills are:
• 32,400 (37.2 per cent) with full skills in Gaelic, that isunderstand, speak, read and write Gaelic;
• 57,600 (66.2 per cent) able to speak Gaelic;
• 6,100 (7.0 per cent) able to read and write but not speakGaelic;
• 23,400 (26.8 per cent) able to understand Gaelic but not speak,read or write it. (National Records of Scotland 2015a)
The 57,600 speakers are mainly resident in the Western Isles, parts of the north
and west mainland and Glasgow. Nationally, 25,000 people aged 3 and over (0.49
per cent of the population) reported using Gaelic at home. The number of speakers
living in the rest of the UK or in other countries is unknown.
45
3.2 Literacy in Gaelic Scotland
Part of the sociology of an orthography is its relationship to who uses it; that is
who learns it and who is literate. This section will review the development of
literacy in Gaelic Scotland to provide the historical context in which the
standardisation of the orthography and the spread of literacy interacted. The role
of literacy in a minority language is complex; on one hand it can confer prestige
and on the other, it can facilitate the acquisition of the dominant language. This
has been attested in other languages (see Grenoble & Whaley 2006: 102-3) and
both elements can be seen at work in the history of Gaelic.
3.2.1 Ogham
The earliest form of Gaelic writing still extant are the stone inscriptions called
Ogham (also known as Ogam). The alphabet is believed to date from the fourth
century and has been found in Ireland, Wales and the Isle of Man in inscriptions
from the fourth to the seventh centuries (Thomson 1994: 220; Ó Baoill 2010: 3).
Examples in Scotland are generally found in the south-west.10 Ogham was also
(arguably) used for non-Gaelic languages in stones in Eastern Scotland in the
eighth and ninth centuries (Ó Baoill 2010: 4). Even after Christianity was
established, bringing with it Latin writing, knowledge of Ogham remained.
Evidence shows it continued to be used (or taught) longer in Scotland than
elsewhere. Damien McManus has argued that despite its distinctiveness in form
from the Latin and Greek alphabets, Ogham has an alphabetic principle which
shows it was derived from an understanding of their systems (McManus 1991).
3.2.2 The Latin Alphabet and Gaelic Script
The foundation stone of modern Gaelic orthography (both Scottish and Irish) was
the adoption of the Latin alphabet. At its arrival in Britain, Latin provided not
simply the physical technology and a writing system but a writing culture and a
foreign language. Ireland was Christianised in the fifth and sixth centuries and
Latin writing spread with the religion (Stifter 2010: 67). The Celtic Church,
brought to Scotland by St Columba (Colum Cille) in the sixth century, developed
the Latin alphabet to form the script used in Latin texts such as The Book of Kells
10 see Forsyth (1996: lxvii) and MacManus (1991) for discussion of these.
46
(written mainly in Iona in the eight century) and the Lindisfarne Gospels (c. 700
from the monastery at Lindisfarne in Northumbria). Due to the adoption of Latin
for written records, there is a scarcity of Scottish Gaelic writing or other evidence
of Gaelic literacy in Scotland. The priority for writing in the Celtic Church was to
create ecclesiastical documents and these were written in Latin (Meek 1993a:
345). Where and when this Latin alphabet began to be systematically adapted for
writing in Gaelic is not clear. However, it is from the beginning that the sound
system of Gaelic had to be adapted to symbols created for different languages and
phonological systems. The decisions about which phonological information needed
to be encoded for the graphic communication to be functional, and which
phonological and phonetic information is unnecessary, began here.
This early period of language, from the adoption of the Latin alphabet to around
900 AD is designated as Old Gaelic or Old Irish (recently scholars have opted to
prefer the former to indicate that its use extended beyond Ireland; see Ó
Maolalaigh (2013b: 42, n. 2). Evidence for Old Gaelic is directly attested in eighth-
and ninth-century glosses and marginalia of Latin manuscripts. At least by the
ninth century in Ireland, Latin had been replaced by Gaelic as the primary writing
system used in monastic schools (Stifter 2010: 67). The literacy training provided
by monasteries and the Church foreshadowed the later importance that churches
had for literacy in the Highlands of Scotland. One of the earliest written Scottish
Gaelic sources is from a Columban monastery in north-east Aberdeenshire. Known
as The Book of Deer, this small Gospel Latin book was produced in the ninth
century and contains two additions in Old Gaelic. The manuscript also contains 5
formal deeds and an origin legend about the monastery of Deer added in the
twelfth century; see Jackson (1972) and the re-assessments in Forsyth (2008). The
use of Gaelic rather than Latin for this demonstrates the value attached by Gaelic-
speaking clerics to formal Gaelic deeds confirming their land-grants and
immunities (Meek 1993b: 238).
On the evidence remaining, it appears that those who used it, a small literary
elite, created a single form of grammar as there is little synchronic variation. At
this time, then, Gaelic writing embodies the idea of a standard language as a
uniform prestige variety. It has been argued that a northern variety of Gaelic was
adopted as the prestige form (McCone 1985) although it is not known if there was
a codified synthesis of dialectal forms but the use of the resulting variety
47
nevertheless shows its prestige (Stifter 2010: 60). Ó Baoill ascribes the
'achievement' to 'the power of both the Church and the secular learned classes' (Ó
Baoill 2010: 5). Despite the consistency in grammar, however, Old Gaelic
orthography is not a fully uniform system (Stifter 2010: 67). David Stifter has
described the use of 18 Latin letters (a, b, c, d, e, f, g, h, i, l, m, n, o, p, r, s, t,
u) as,
Especially unsuited for rendering the phonemic system of Old Irish withits more than sixty phonemes (including diphthongs). This means thateach letter has to bear the functional load of expressing around fourdifferent phonemes. (Stifter 2010: 67)
The system created, therefore, elaborated on the restricted alphabet by using
word position and giving diacritic functions to some letters along with their
phonemic value as well as diacritics themselves, in particular the punctum delens
(the placement of a dot above a letter to indicate lenition).
Gaelic was written in a modified version of the Latin script which combines
features of cursive and half-uncial Roman script. It is known as Insular miniscule
or an cló Gaelach. One of the early pieces of evidence of it is in Cathach Choluim
Chille, a psalter from the second half of the sixth century (Bannerman 1983: 214).
It was used in Ireland for writing both Irish and Latin. Over 100 manuscripts
between the ninth century and the second half of the sixteenth century were
written in clò Gaelach. It was used in print in Ireland from the seventeenth
century. Gaelic type did vary between then and the twentieth century, but 'in
spite of variations these were all akin to one another and clearly distinguishable
from Roman type' (Ó Cuív 1969: 24-25). Although, when the complete translation
of the Bible into Irish Gaelic was published in Ireland in 1685, the cló Gaelach
script appears to have become unfamiliar enough to Scottish Gaelic readers that
the Rev. Robert Kirk (?1644-92) transcribed it into Roman print.
3.2.3 Middle Gaelic and Classical Gaelic
The Middle Gaelic period, traditionally dated from c.900 AD to c. 1200 AD, does
not refer to a uniform variety, rather the period covering the transition between
Old and Modern Gaelic. The end of the Middle Gaelic witnessed what Ó Cuív (1980)
has referred to as a 'mediaeval exercise in language planning', which resulted in
48
the codification of language for use particularly in the composition of syllabic
verse, a form of language which was propagated and disseminated in the Early
Modern bardic schools. Linguistic development and divergence in Scotland
independent from Irish Gaelic is likely to have begun long before this (Ó Maolalaigh
2008); however, concrete evidence from this period is lacking (Gillies 1993: 145)
and we need to rely for the most part on linguistic reconstruction (Ó Maolalaigh
2008). Despite these developments the writing system itself made gradual changes
rather than undergoing total reform:
[the writing system] remains remarkably stable until the end of thetwelfth century (...) The main deviations from the earlier standard liein frequent spelling confusions of those sounds that had merged, e.g.nd and nn, or oí/aí/uí, etc. (Stifter 2010: 111)
The examples given by Stifter demonstrate how sound changes inevitably lead to
graphic changes. Once the different forms of <nd> and <nn>, for example, are no
longer needed to encode phonological data the choice of writers to retain them is
for other reasons including historical continuity and conservatism.
The written language from around 1200 to 1600 is known as Classical Gaelic. It
was used in Ireland and Scotland as a formal written convention. This convention
masks dialectal divergences between Irish and Scottish Gaelic over this period (Ó
Maolalaigh (1998: 12-16). Although Gaelic at the time ceased to be used by the
Scottish royal court, being replaced by Scots in the court and in burghs, areas
outwith the political reach of the court were able to maintain different traditions.
In the areas controlled by the Lordship of the Isles, principally the West Highlands,
the Lords encouraged what Donald Meek has referred to as a Gaelic 'civil service';
clerics, judges, genealogists, sculptors and physicians. This literate class were
capable of using Classical Gaelic (Meek 1993: 345). There is little to suggest that
literacy went beyond this élite class, or that there was a desire to extend it beyond
them. The strongly conservative and prescriptive nature of Classical Gaelic is
summarised by Colm Ó Baoill,
This learned language remained the rigid and unchanging standard forprofessional poets in Scotland and Ireland down to the seventeenthcentury, becoming further and further removed from any spoken dialectas the centuries went on. (Ó Baoill 2010: 11)
49
The conventions of Classical Gaelic were not disrupted by the Reformation. It is
believed that the tradition of writing Classical Gaelic was carried on by priests
who had trained in bardic schools who subsequently became ministers (Meek 1993:
345).
Although Classical Gaelic was used and understood as a standard in the Early
Modern period, it was not an utterly inflexible and invariable one. Thomson finds
that 'at all times this language is open to modification in the direction of spoken
Scottish Gaelic' and the same modifications in Ireland to spoken Irish (Thomson
1994: 99). For writers in the twentieth century who are reflecting on the patterns
of use in early centuries, it can be easy to judge them in the frame of the ideology
of the standard and the expectations of modern readers. Thus a lack of consistency
becomes 'notorious' (Thomson 1994: 99). Donald MacKinnon noted that in the early
MSS 'authors and scribes were not so much impressed with the importance of
uniformity in orthography as we have become' (1909: 3).
A significant break from the Classical orthographic tradition is found in one of the
major manuscripts of the sixteenth century. The Book of the Dean of Lismore
remains one of the most important Scottish collections of Gaelic poetry. It was
compiled between 1512 and 1542, by the Dean of Lismore, James MacGregor, and
his brother Duncan, but with a spelling system based on Middle Scots and written
in Scots 'secretary hand' (Meek 1993: 345; Meek 1989: 132; Ó Baoill 2010: 14). Its
unconventional spelling was undoubtedly influenced by geography, as its scribes
were located in Fortingall, Perthshire, far from the West Highlands where the
Lords of the Isles and the conventions of the bardic schools in Gaelic writing were
in control.
The Highland historian John Bannerman considers the sixteenth century as 'the
earliest for which a seriously researched statement as to the incidence of literacy
in the Highlands can be made' (Bannerman 1983: 214). Even then, literacy appears
to be very restricted in spread. It is only in the second half of the sixteenth century
that there is a noticeable increase in the numbers of Gaelic speakers able at least
to append their signatures in Scots to legal documents (Bannerman 1983: 216).
For a recent discussion of where and how Gaelic was written in late medieval and
early modern Scotland, see MacCoinnich (2008).
50
3.2.4 The Seventeenth Century
The influence of the bardic schools lessened as power waned in Gaelic-speaking
territories. In 1609, the Statutes of Iona enacted by King James VI of Scotland
(1567-1625) included forcing the education of the clan chiefs' heirs in the Lowlands
and outlawing the patronage of Gaelic poets. The spread of Gaelic literacy was
also threatened by the efforts to remove Gaelic completely. In 1616, the Act for
the Settling of Parochial Schools explicitly claimed that an aim of it would be
that the vulgar Inglishe toung be universallie plantit, and the Irishelanguage, whilk is one of the cheif and principall causis of thecontinewance of barbaritie and incivilitie amongis the inhabitantis ofthe Ilis and Heylandis, may be abolisheit and removeit (quoted inDonaldson 1970: 174-75)
Outside of printed materials, we know some of the scribes who would have been
active in the Highlands. They were literate in Latin and in Scots, which was the
language of government. Toirdhealbhach Ó Muirgheasáin, one of two scribes for
Ruairi MacLeod of Dunvegan, drew up and signed a contract in Scots in 1614
(Bannerman 1983: 217-8). The National Archives of Scotland holds a contract of
fosterage he wrote in Classical Gaelic (Thompson 1994: 220). The Classical Gaelic
poet and scribe for the MacDonalds of Clan Ranald, Cathal MacMhuirich also wrote
and signed documents in Scots (Bannerman 1983: 218; Thomson 1994: 185;
MacCoinnich 2008). These point to the fact that literate Gaelic speakers, at this
early stage, are multilingual and were as familiar with writing systems of other
languages as they were with Gaelic.
The Church of Scotland's Synod of Argyll was a 'relatively Gaelic-friendly'
institution (Ó Baoill 2010: 14). Ministers of the Synod of Argyll completed a
translation of the Old Testament by 1673 although the manuscript does not survive
(Meek 1997: 16). It had more success with other religious works including the
Shorter Catechism or Adtimchiol an Chreidimh (1631) translated by Rev. Ewen
Cameron of Dunoon. Only the second edition from 1659 survives (Thomson 1962).
It contains examples of usage which appear to be Scottish Gaelic and shows a
move away from Classical Gaelic.
51
3.2.5 The Eighteenth Century
The eighteenth century is considered to be the key period for the development of
a Scottish Gaelic orthography based on the contemporary vernacular. As MacBain
describes it in his 'Outlines of Gaelic Etymology' with which he prefaces his
dictionary:
In the eighteenth century Scottish Gaelic broke completely with theIrish and began a literary career of its own with a literary dialect thatcould be understood easily all over the Highlands and Isles. (MacBain1911: iv)
At the start of the eighteenth century, Edward Lhuyd commented on the disparity
that had been created between Classical Gaelic writing and the vernacular:
the ancient Orthography whereof has been much better preserv'd thanits Pronunciation [...]; Pronouncing as different from what they write,as the French. (Lhuyd 1707: 2)
As the century continued, however, the grammar and vocabulary of the written
language was reformed to the vernacular.
Who forms the usage climate, in an age lacking grammars, dictionaries,and other linguistic resources? In a word: authors. Literary authors, inthe first instance; academic authors, in the second. (Crystal 2006: 55)
As David Crystal observes above, authors are the main agents of linguistic
authority in a written culture without grammars and dictionaries. In eighteenth-
century Scotland, it was Gaelic authors, religious in the main, who formed the
usage climate. One of the key figures in this period was the poet and teacher
Alasdair Mac Mhaighstir Alasdair (Alexander MacDonald); considered by Ronald
Black as one of the 'founding fathers' of modern Gaelic orthography (Black 2010:
237). As a literary author, he authored the first published secular book with his
collection of poems, Ais-éiridh na Sean Chánoin Albannaich (Mac-Dhonuill 1751).
As a putative lexicographer, his published wordlist, Leabhar a Theasgasc
Ainminnin: A Galick and English Vocabulary (M'Donald 1741) came under the
auspices of a religious organisation.
The authors with the most impact on the usage climate, however, were the
translators of the Bible into Scottish Gaelic. The influence of southern dialects, a
52
point that would be contentious later on, began with these authors. The work of
the translation of the Bible into Scottish Gaelic has been documented and
described in depth by Meek (see Meek 1997). The translation took place between
1755 and 1801, the New Testament appearing first in print in 1767. The different
books of the Old Testament took longer to appear and the whole work was
completed in 1801.
Towards the end of the eighteenth century there was a growth in the linguistic
study of Gaelic following on from the sporadic works of Lhuyd and MacMhaighstir
Alasdair's wordlist. Robert MacFarlan was the first and only ‘Professor of Gaelic’
appointed by the Highland Society (Gillies & Pike 2012: 207). He published a
dictionary, Nuadh Fhoclair Gaidhlig agus Beurla [New Gaelic and English
Vocabulary] in 1795 dedicated to the Society’s members. In it, he expresses his
dismay at the decline of Gaelic and shows his concern at the lack of a standard
Gaelic orthography. The Rev. William Shaw also published a grammar in 1778 and
a dictionary in 1780 for which he gathered some material orally on a trip to the
Highlands in 1778 (Gillies & Pike 2012: 206). In contrast, he is clear that he is not
attempting any orthographic revision in his dictionary:
The Gaelic reader will find no innovations in orthography; for I haveconsidered it my business rather to record words as they have beenwritten in the ancient Irish MSS. than attempt to write a Dictionary, byaltering the spelling from the received method, to what I mightconceive it ought to be, according to the powers of the letters, and thephilosophy of the language. (Shaw 1780: 8)
Literacy
In the formal educational gap left by the state, evangelical societies attempted
to bring literacy, and with it the word of God, to the Highlands. The most
signficant of these was the Society in Scotland for the Propagation of Christian
Knowledge (SSPCK) founded in 1709 to spread Protestant Christianity. This
evangelical impetus, along with the Gaelic Bible as a principal text-book, meant
that literacy went hand in hand with Christianity. Again, this is not unique to the
spread of Scottish Gaelic literacy:
One thing seems clearer, however, namely that vernacular literacy perse is usually merely an intermediary goal along the road to something
53
other and greater than itself, something which (it is hoped) literacywould enable the literate to attain (Fishman 2010: 84)
In this case, that which was 'greater than itself' was access to the word of God.
The SSPCK were initially no more open to Gaelic than the writers of the 1616 Act
and over the decades Gaelic's association with 'barbaritie and incivilitie' was still
strong, with the added negative factors of Jacobitism and Catholicism as this
extract from the SSPCK records show:
Another unfavourable circumstance is the Erse tongue. (...) a numberof arrful<sic> and bigotted priests, men who 'compass sea and land togain one proselyte,' are either born to the knowledge of their language,or spare no pains to acquire it. Indeed there are no Erse Grammars butthose that are printed at Rome, and taught at some of the Colleges inFrance. (SSPCK 1771: 3)
SSPCK teachers were instructed not to use Gaelic in their schools. However, before
the end of the eighteenth century, the Society had been convinced that teaching
monolingual Gaelic speakers literacy in their own language first would be effective
in later teaching them English.
Durkacz characterises the Church's position towards Gaelic in eighteenth century
as ambivalent; supporting 'English in education, serving the long term aims of
civilising and reforming the Highlands, and Gaelic in preaching and worship (...)
holding back the counter-Reformation' (Durkacz 1981: 147). The European
creation of mass literacy is closely linked to the establishment of 'national
languages' understood as 'codified written languages suitable to unite those
literate in it to form a national community' (Coulmas & Guerini 2012: 438). The
expansion of Gaelic literacy in the eighteenth century took place in this context,
but it had to work against the British project of English as the 'national language'.
Scottish Gaelic, moreover, was associated with rebellion and Catholicism and the
Church regarded the Anglicisation of Gaelic-speaking areas as an essential aspect
of reformation (Durkacz 1981: 147).
While mass literacy was yet to be achieved in the eighteenth century, the literate
class of Gaels and Highlanders, particularly ministers, had both literacy and access
to the written tradition of Gaelic. In the Rev. Thomas M. Murchison's writings on
church history in the Highlands, he cautions against underestimating the
availability of books. As an example, he details the library of a Rev. Donald
54
MacKilligan, minister of Alness, who in 1715 had an English Bible, an (Irish) Gaelic
Bible, a Greek New Testament, a Latin grammar, several theological works,
Woodridge's book on gardening and a book by Lord Gray on equine health
(MacCalmain 2010: 91). There were also presbytery libraries in places such as
Glenelg.
The eighteenth century saw a period of reform in the orthography of Scottish
Gaelic as writers sought to vernacularise the Classical Gaelic norms to fit the
modern language as spoken in Scotland. This was in part motivated by religious
and political desires to bring religion, and 'civilisation', to the Highlands.
3.2.6 The Nineteenth century
The nineteenth century continued the influence of ministers as arbiters of
linguistic usage and in various roles beyond translation: as magazine publishers,
as dictionary makers and as grammarians. The aims of literacy were the same for
the early part of the century as the eighteenth century; Gaelic literacy was a path
to God. There are accounts of successful acquisition of literacy, although these
include accounts of literacy as a path to English.
Linguistic works
The first linguistic book of the nineteenth century was Elements of Galic Grammar
(1801), the grammar by the Rev. Dr Alexander Stewart (1764-1821). Stewart was
a son of the Manse of Blair Atholl and became a minister himself at Dingwall and
later at the Canongate, Edinburgh. He later revised the SSPCK translation of the
Scriptures. It was while he was minister at Moulin, Perthshire that he first
published his Elements. It remained influential throughout the century, with five
editions between 1801 and 1901. In 1801, Stewart saw the orthography as a work
in progress, promising to 'assist the reader in forming a judgment of its merit, and
how far it may admit of improvement' (1801: 30). In the first edition Stewart calls
for editors and writers to use h with aspirated l, n, r, confident that the change
would be welcomed. But by the second edition, Stewart's hopes of altering the
written form of aspirated l, n, and r, have changed:
55
It is perhaps too late, however, to urge now even so slight an alterationas this in the Orthography of the Gaelic, which ought rather to be heldas fixed beyond the reach of innovation, by the happy diffusion of theGaelic Scriptures over the Highlands. (Stewart 1812: 31)
The Rev. Alexander Stewart, along with the Rev. Alexander Irvine, had a direct
influence on the form of orthography used in the spread of literacy: Alexander
MacLaurin credits them with revising his First Book for Children, clearly appealing
to their recognised authority.11
The manuscript having been submitted to the revision of the Rev. MrStewart of Dingwall, and the Rev. Mr Irvine of Little Dunkeld, wasafterwards presented by Mr M'Laurin to “The Society for the support ofGaelic Schools,” and is now printed for the use of the Schools undertheir charge. (M'Laurin 1811: 1)
In the 1820s, there was a series of argumentative letters in the Inverness Journal
between the Rev. Dr Alexander Irvine and the Rev. Dr Thomas Ross about the
translation and orthography of the Gaelic Bible which revealed the tensions
between dialects. Ross claimed that:
One of the greatest objections to that translation is, that it is intoGaelic of a much too local nature, - that it is peculiarly adapted to theHighlands of Perthshire, while it is but partially intelligible in otherparts of the Highlands. (Ross 1821: 13)
In response, the Rev. Dr Irvine, argued that the Bible was, in fact, supradialectal,
differing from all dialects:
(…) till the days of Dr Ross (…) the merit of the translation was neveronce called in question, though it was not faultless, and though it mightdiffer from the dialectal variations of language in Ross, Inverness,Argyle, and Perthshire. For it differs from all; and this is just one of itsexcellencies. (Ross 1821: 25)
The Rev. Alexander Irvine (1773-1824) was a Church of Scotland minister and
poetry collector from Fortingall, Perthshire. The son of a tenant farmer, he
attended the Universities of St Andrews and Edinburgh and ministered in Mull,
Rannoch and Little Dunkeld, Perthshire (Hogg 2011). While he did not author many
Gaelic texts, he revised and prepared for publication the quarto edition of the
11 Although MacLaurin doesn't follow their conventions on the use of the accent. See 8.1.
56
Gaelic Bible issued by the SSPCK and assisted in the Dictionarium Scoto-Celticum
(1828) produced by the Highland Society of Scotland (Scott 1923: 159). He also
collected Ossianic poetry, the MS of which is now held at the National Library of
Scotland. A monument to him was erected by subscription in the parish church at
Little Dunkeld (Edinburgh Magazine 1825: 381).
An early dictionary was Peter MacFarlane's Focalair Ur Gaelig agus Beurla (1815).
MacFarlane (1753-1832) was a schoolmaster in Appin and he translated religious
works. His dictionary has a dialectal bias towards Argyllshire Gaelic (Gillies & Pike
2012: 209). He included a short section on 'General Rules for Reading the Gaelic
Language', specifying that they are mainly taken from the Gaelic Bible. Robert
Armstrong (1788-1867) produced A Gaelic Dictionary in 1825. Originally from
Perthshire, he studied at St Andrews University and worked as a schoolmaster in
London (Gillies & Pike 2012: 210) His dedication to King George IV preceded his
appointment as Gaelic Lexicographer in Ordinary, and Gillies & Pike suggest that
he benefitted from a pro-Scottish mood after the King's famous visit to Edinburgh
in 1822 (2012: 210). He had an etymological interest and was concerned with the
antiquity of the Celtic languages (Gillies & Pike 2012: 211).
The Highland Society of Scotland was founded in 1784 to promote the interests
and culture of the Highlands and Islands. In 1828, it published its Dictionarium
Scoto-Celticum. It was the first Gaelic dictionary to be compiled by a team; the
Rev. Dr John Macleod, minister of Dundonald, assisted by Ewen MacLachlan, the
Rev. Dr Alexander Irvine of Little Dunkeld, and the Rev. Alexander Macdonald of
Crieff. The overseer of the printing of the text was the Rev. Mackintosh Mackay,
minister of Laggan (Gillies & Pike 2012: 212). Its prefacing grammar is an abridged
version of Stewart (1801). Its spelling system was based on that of the scholars of
the SSPCK bibles who it credited with creating a standard orthography.
This deficiency was happily and in a great measure, supplied by thetranslation of the Scriptures, and the publication of them in Scoto-Gaelic, by the Society in Scotland for Propagating Christian Knowledge.The system of orthography followed there, adopted, as it was, bynatives of intelligence and learning in the Scottish Highlands, andimproved by successive editions of the Scriptures, has been strictlyadhered to in the present work. (HSD 1828: xi)
57
It dealt with orthographic variation by including entries for variants that cross-
referenced back to the main entry.
The Rev. Dr Norman MacLeod (1783-1862), known as 'Caraid nan Gàidheal', The
Friend of the Gaels, was from Morvern and was a major figure in Gaelic literary
production in the nineteenth century as well as developing education in the
Highlands and organising relief during the potato famines in the 1830s and 1840s
(see Gillies & Pike 2012: 216-7). The Rev. Dr Daniel Dewar (1788-1867) was born
in Glen Dochart, Perthshire and studied at Edinburgh University (see Gillies & Pike
2012: 216-7). MacLeod and Dewar's dictionary aimed to fill the need of a small,
moderately priced dictionary that condensed the existing lexicographical work
done by Armstrong and the Highland Society. It was first published in 1830, with
a second appearing in 1832, and twelve more between 1833 and 1910.
Neil MacAlpine (1786-1867) was a native of Islay and a schoolmaster there. In 1832,
he published The Argyllshire Pronouncing Gaelic Dictionary in which he included
a substantial 'Rudiments of Gaelic Grammar'. Intended for use in Gaelic schools,
it drew on Armstrong (1825) and the Highland Society Dictionary (1828) with his
own additions from Islay usage and other sources including Lewis, Skye, Lochaber,
Harris, and Perthshire usages (Thomson 2004). Its usefulness and impact is
reflected by the fact that it was reprinted fourteen times between 1833 and 1973.
John MacKenzie's An English-Gaelic Dictionary was written in 1847 to specifically
accompany MacAlpine's Gaelic-English work. John MacKenzie (1806-1848) was born
in Gairloch and worked as a joiner, debt collector and book-keeper before taking
up a job in Edinburgh with the MacLachlan and Stewart publishing house (Gillies
& Pike 2012: 221). He published several collections of poems, an edition of
MacMhaighstir Alasdair's poetry (1834) and a history of the 1745 rebellion
Eachdraidh a' Phrionnsa [The Prince's History] (Mac-Choinnich 1844). It was his
Sàr-Obair nam Bàrd Gàelach [The Beauties of Gaelic Poetry] (1841), a major
anthology of Gaelic poety, that became his calling-card. Introducing it, MacKenzie
acknowledges that his edition will have 'a few derivations from what is generally
recognised as the standard of Gaëlic orthography' and in order that he can 'do
justice to the harmony of the versification, no acknowledged rules will apply'
(MacKenzie 1841: iv).
58
Alasdair Roberts has argued for Father Ewen MacEachen as a 'neglected pioneer'
of Scottish Gaelic orthography (Roberts 2006: 365). Father Ewan, or Evan,
MacEachen [Eobhan MacEachainn] (1769-1849) was a native of Arisaig in the West
Highlands (Robert 2006: 358). He trained as a priest in Valladolid, Spain, before
returning to the Highlands where he built the first chapel in Laggan and practised
for twenty years in Braemar (MacEachen 1902: iv). He translated several religious
works into Gaelic including An Cath Spioradail (1835), a translation into Gaelic of
Il Combattimento Spirituale by Dom Lorenzo Scupoli, and Leanmhuinn Chriosta
(1836), a translation of The Imitation of Christ by Thomas à Kempis (Roberts 2006:
360). His translation of the New Testament, was also revised by the Rev. Colin
Grant, Strathglass, afterwards Bishop of Aberdeen, and published in 1875 (Grant
1875).
Frustrated by publishing errors and rows with other bishops over his translations,
MacEachen published his own dictionary in 1842 (Roberts 2006: 378).12 Although it
appears to have had little impact at the time, it was published in four revised
editions in the early twentieth century – after extensive orthographic revision by
John Whyte and Alexander MacBain – and became the recommended schools
dictionary. MacEachen's original dictionary was a response to the written
conventions of the time. He was keen to label his dictionary as containing ‘pure
Gaelic words’ in contrast to the errors he saw in existing bibles (MacEachen 1842:
iii). He prioritised his own dialect of Arisaig and stated that vernacular speech was
his guide:
in Gaelic authors no improvement can be expected unless they throwaside the Bible and take their expressions, as I have done, from themouth of the best Gaelic speakers. (MacEachen 1842: iv)
He regularised the spelling he found in MacLeod and Dewar. One of his conventions
was to use str word initially replacing existing spellings with sr-. This was undone
in the second edition of this dictionary, revised by MacBain and Whyte (MacEachen
1902: iv).
12 The title page was erroneously dated 1862 (Roberts 2006: 365)
59
Literacy in the nineteenth century
The vital period for widespread literacy was the beginning of the nineteenth
century, with the establishment of the Gaelic Schools Societies from 1810 and the
completion of the translation of the Scriptures. The Edinburgh Society for the
Support of Gaelic Schools was set up in 1810. Together with Alexander MacLaurin,
the Rev. Christopher Anderson, who founded the Society, produced readers and
textbooks in Gaelic (e.g. MacLaurin 1811: 1816). A year later, a similar Glasgow
society was founded and another in Inverness in 1818. The Edinburgh society
taught Gaelic only and those of Glasgow and Inverness taught writing, counting
and English.
The relative success of the effort to spread Gaelic literacy in the nineteenth
century should not be underestimated. The editor of the periodical An Teachdaire
Ùr Gàidhealach claimed a twenty-fold increase in literate Gaels between 1829 and
1835.13 There is evidence of Gaelic literacy in even the most remote areas. In An
Teachdaire Gàidhealach, April 1857, there is an account of a visit to 'Eilean Irt',
presumably Hiort, St Kilda, from a Mr. Mac Dhaibhidh na h-Earradh. He says he
visited on Tuesday 2 September 1856 on board the factor's ship the Cornwallis for
communion. He visits the school, taught by a Mr Kennedy and describes the
abilities of the pupils:
Bha cuig deug thar fhichead 's 'a sgoil. Bha dha-dheug dhiubh a leughadhBeurla, ach cha robh leabhraichean freagarach aca; bha ceithir deug asgirobhadh<sic> agus naoidh thar fhichead a leughadh Gailig. (MacDhaibhidh na h-Earradh 1857: 6)
There were 35 in the school. Twelve of them were reading English, butthey didn't have suitable books; fourteen of them were writing andtwenty nine reading Gaelic.
The spread of evangelical Presbyterianism along with Gaelic literacy has been
credited by Charles Withers as strengthening Gaelic in some areas, but also
13 ‘Tha nis anns a’ Ghàidhealtachd fichead a leughas agus a sgrìobhas a’ Ghàilig air son a h-aon abh’ ann ’nuair thòisich an seann Teachdair’ air dol a mach.’ (MacLean 1835: 2) In the Highlandsthere are now twenty who read and write Gaelic for every one there was when the oldMessenger was going out.
60
establishing a diglossic situation. In his assessment, Highland schools teaching
Gaelic scriptures:
not only helped to reinforce Gaelic's place as a church language butfurther influenced Highland consciousness in making a distinctionbetween Gaelic as the language of spiritual affairs, English of worldlyadvance. (Withers 1988: 151)
In Grenoble & Whaley's guide to language revitalisation Saving Languages (2006),
they acknowledge that the role of literacy in language revitalization is complex as
it can have positive effects such as prestige and negative ones such as facilitating
acquisition of a majority language (Grenoble & Whaley 2006: 102-3). One of the
arguments made to promote the teaching of Gaelic literacy first in schools was
that it would encourage speakers to advance to reading, and learning, English.
Accounts from the time appear to give evidence for this. The Rev. Dr Macleod
observed in the Gaelic Schools Society's annual report in 1833:
English reading, and English speaking have made greater progress in theHighlands and Isles of Scotland, since the system of Gaelic teaching hasbeen acted upon – that is, during the last twenty years – than it did forcenturies before then. (quoted Withers 1984: 148)
Personal accounts of Gaelic literacy being a path to English can also be found such
as this one from the Gaelic periodical An Teachdaire Gaidhealach published in
Australia:
An deigh dhomh comas fhaighinn air a Ghaelic a leughadh, cha'nfhoghnadh so leam gu'n oidheirp a thoirt air a' Bheurla; cha robh moranleabhraichean Gaelic 'ra fhaotainn.' ('An Ceannaiche Og Gaedhealach'1857: 6)
After I was able to read Gaelic, I wasn't satisfied without making aneffort to learn English; there weren't many Gaelic books available.
3.2.7 The Twentieth Century
By the start of the twentieth century, some provision was beginning to be made
for the teaching of Gaelic in schools. This revived the more pragmatic educational
motives for uniformity and simplicity in order to achieve literacy. MacBain and
Whyte's third edition of MacEachen's Dictionary (1906 [1842]) was recommended
61
by the Education Department as an orthographic guide for the Leaving Certificate
Examination in Gaelic (Gillies & Pike 2012: 227).
Alexander MacBain (1855-1907) was from Badenoch in the North Highlands
(MacBain 1911: vi). The son of a farm worker, he worked for the Ordnance Survey
in Scotland and Wales before attending King's College, Aberdeen, and graduating
MA with honours in philosophy. He was rector at Raining's school, Inverness, until
his death. An active participant in the Gaelic Society of Inverness he wrote many
papers on Gaelic philology and edited the Celtic Magazine and Highland Monthly.
He published in 1896 the first Etymological Dictionary of the Gaelic Language. The
first edition was sold out within a year and a revised edition was published in 1911.
One of the key influential linguistic works appeared in the first decade of the
century. Edward Dwelly (1864-1939) was a native of Twickenham, Middlesex and
learner of Gaelic. He was a soldier with the Argyll and Sutherland Highlanders,
then the Seaforth Highlanders and he married a native Gaelic speaker from
Ardchattan in Argyll in 1896 (Gillies & Pike 2012: 227). The dictionary he created
appeared in instalments between 1901 and 1911. The complete dictionary was
reprinted in a three volume set in 1911 and he was awarded a pension by King
Edward (Gillies & Pike 2012: 234). His dictionary did not provide etymological
information, and along with existing texts, he collected words from a network of
informants. From this, it appears his interest was with the language as it existed,
rather than its philological and antiquarian value. His legacy was to have created
what would be the lexicographical authority in Scottish Gaelic of the twentieth
century.
Other dictionaries produced in the twentieth century showed an interest in
collecting dialectal variation. The Rev. Malcolm MacLennan (1862-1931) from
Cnip, Lewis revised the Gaelic Bible and published books of hymns, and edited An
Deò-ghrèine between 1906 and 1908. His connections to the Western Isles and
other scholars from the North-West Highlands and Islands meant that his dictionary
was the first such to record many Hebridean words (Gillies & Pike 2012: 238).
Heinrich (Henry) Cyril Dieckhoff (1869-1950) was a German phonetician and a
priest at Fort Augustus abbey. In 1932, he published A Pronouncing Dictionary of
Scottish Gaelic, based on the Glengarry dialect. His motivation was the
preservation of the traditional pronuncation of older speakers.
62
3.2.8 The GOC Era
In 1976, the Scottish Certificate of Education Examination Board (SCEEB) (later
the Scottish Qualifications Authority (SQA)) set up, at the request of their Gaelic
Panel, a sub-committee to draw up guidelines on Gaelic spelling. Donald MacAulay
describes the impetus as coming from the dissatisfaction of school examiners with
pupils' literacy:
the Scottish Certificate of Education examiners have been so unhappywith the problem that they have prevailed on the Examination Board'sGaelic Panel to establish a committee to look into the matter.(MacAulay 1976: 85)
The first report, entitled Gaelic Orthographic Conventions and known as GOC, was
published in 1981. It was intended to be used in the SCEEB's exam questions from
1985 and used by candidates from 1988. The remit was not to radically reform or
remake the orthography but to produce a set of standard orthographic conventions
which would be utilised by the Board's examiners, setters and markers as basic
criteria in dealing with Gaelic examination papers and scripts and would serve as
a guide to teachers and to candidates preparing for the examinations.
The existence of the report provided a benchmark against which subsequent
publications defined themselves; editors and authors often explaining if they were
following GOC and what exceptions if any they were choosing to make. The GOC
reforms can be described as largely successful in the sense that they have been
widely disseminated and adopted – at least recognised as an authority if not
carefully followed in practice.
In 1982, the Gaelic Books Council (at that time based at the Department of Celtic,
University of Glasgow and know in Gaelic as An Comunn Leabhraichean), published
a booklet entitled Facal air an Fhacal covering publishing-related news. It included
an article, in English, on the 'New Spelling' which the issue itself largely follows.
The article notes that:
it seemed worthwhile, in this first issue of Facal air an Fhacal, toprovide some information on the report which recommends the changesand about which very little seems to be known outside educationalcircles (Anon. 1982: 23)
63
The article uses quote marks for 'new spelling' as it subsequently reassures readers
that:
some of the conventions recommended are widespread already: forexample seo and siud for so and sud and the use of <a> for unstressedvowels which is already quite well established, although far fromuniversal (Anon 1982: 23-4).
The recommendations highlighted in the article are those where one form is
preferred over variety, and the wariness of orthographic diversity is seen in its
description of GOC's convention on the use of hyphens:
Hyphens are introduced in 'certain expressions which are strongly feltto constitute a unit but have primary stress on a non-initial element',and the adoption of this recommendation would eliminate theconfusing diversity which obtains at present. (Anon 1982: 24) boldadded
In the early 1990s, it was felt that the orginal report required extending. A sub-
committe of the Gaelic Panel of the Scottish Examination Board was created,
convened by Ronald Black, University of Edinburgh, Kenneth MacDonald,
University of Glasgow, Iain MacIlleChiar, Northern College of Education and
Rosemary Ward, Strathclyde Regional Council. However, the Principal Examiners
for Gaelic responded that the changes recommended were too substantial when
familiarity with GOC 1981 was recent. They strongly rejected many of the points
made and criticised what they considered a lack of consultation, believing that
the proposals would not be accepted by teaching staff. The report was not
published.
The second publication in 2005 superseded the first GOC report. The brief for the
panel carrying out the work was that 'the document should be reviewed and
updated, whilst adhering to the principles and recommendations contained in the
original' (SQA 2005: 1). The Project Consultant was Donald John Maciver,
Comhairle nan Eilean Siar. The steering group comprised Annie MacSween, Lews
Castle College, Boyd Robertson, University of Strathclyde and Ian MacDonald of
the Gaelic Books Council. A third edition of GOC was published in October 2009
with the same steering group and project consultant. This third edition made
minor amendments to the 2005 report.
64
4 Ideologies in Gaelic Spelling
Pleine de cédilles et de fureur, l'histoire de l'orthographe française estavant tout une bataille des idées. (Cerquiglini 2004: 7)
Full of cedillas and fury, the history of French orthography is above alla battle of ideas.
This chapter asks which are the language ideologies that have been at work in the
standardisation of modern Scottish Gaelic orthography. Beliefs about language,
known as language ideologies, shape all aspects of how people use language and
talk about language. They include beliefs about how written language is spelled,
how it should be spelled and who influences and manages the application of these
beliefs to shape the creation of the standard. The ideas are often at odds with
one another. In this way, the comment made by Cerquiglini about French
orthography above is also true in this case: the history of Scottish Gaelic
orthography is a battle of ideas and the path towards compromise between the
competing ideologies is difficult. Thus Black refers to the reform of Classical
Gaelic to vernacular Scottish Gaelic as 'the long and painful process of adaptation'
(Black 2008: 74).
There are several beliefs and narratives that illustrate language ideologies in
relation to the development and standardisation of Scottish Gaelic spellings. The
motivations that underlie these ideologies are the need to defend Gaelic as a valid
language and the desire to maintain Gaelic as a living language.
This chapter considers the following: what does 'standard' mean in terms of Gaelic
spelling; what role do the phonographic system and the phonographic ideal have
in Scottish Gaelic standardisation; what place does variation have within spelling
conventions; what place do historicity and etymology play; and how does
orthographic standardisation relate to language status.
4.1 'Standard' Language in the Scottish Gaelic context
As was discussed in Chapter 2, having a standard confers social prestige on a
language; it is a mark of a civilised and fully developed language and, therefore,
defends it and its speakers from those seeking to denigrate it. However, 'standard'
can be understood in different ways. It can be a codified prestige variety based
65
on that of high-status speakers such as in English or French. It can have the
meaning of the 'correct' form of language as in Milroy's definition of the ideology
of the standard. Or it can mean a set of conventions used to facilitate interaction,
in the same way as standard weights and measures.
The popular European understanding of standard languages is that they are the
models of correct language that are based on rules. 'Standard' spelling in Gaelic is
understood in different ways. The translation of the terms standard and
conventions into Gaelic can reveal writers’ attitudes as to their meaning. In the
early nineteenth century, the idea of the standard as correct language is evident
in the use of the term ceart-sgrìobhadh, 'correct writing', to mean spelling or
orthography. This is found in several dictionaries including; Shaw (1780) which has
Ceartſgriobham for 'spell', MacLeod & Dewar (1831) which has both litirich, ceart-
sgrìobh for 'spell' and Dwelly (1911) which has ceart-sgrìobhadh and litreachadh
for 'orthography'. The online dictionary LearnGaelic.net also currently includes
ceart-sgrìobhadh as 1. correct writing 2. orthography. The only example so far
located of ceart-sgrìobhadh in a text other than a dictionary is in An Teachdaire
Gàidhealach (Australia) where it is used but also glossed in English for the reader
as 'spelling' ('Naigheachdan' 1857b: 8).
The use of litreachadh for 'spelling', from litir, 'letter', is more frequent in the
late nineteenth and twentieth century. It is used in the corpus, for example, by
Archibald Sinclair in An t-Òranaiche (1879), and by Donald MacKinnon in his
introduction to Am Fear-Ciùil (1910). It continues to be the most common
translation of spelling in the twenty-first century, used by Cox in Ri Linn nan
Linntean (2005), along with modh-litreachaidh, 'way of writing' (2005: xii). In
Saoghal Bana-mharaiche (2007) litreachadh, 'spelling', is also contrasted with
nòsan, 'conventions'.
Orthography is also regularly interpreted as 'rules'. In their dictionary, MacLeod &
Dewar say they adhere to the 'rules' prescribed by the Gaelic Bible (MacLeod &
Dewar (1831: vi). MacDonald complains that varied usage gives the impression that
no-one ever laid down any riaghailt, 'rule' (MacDonald 1875: 118). Despite the GOC
committee's choice of the word conventions and the Gaelic translation of
gnàthachas to describe their report contents, as in other spelling reform contexts,
the prescriptive intent led some to the metaphor of legalistic language and 'rules',
66
for example Donald Meek, who described GOC in a radio interview as riaghaltan
ùra, 'new rules' (Dealan-dè 1986). In the next example, the legalistic metaphor
'ruling' is used to describe the recommendations:
Initially accents sloping both right and left were used, however theruling now allows only those sloping in one direction. (Garvie 1999: 7)
The title of GOC is translated as such by the author in this example (Riaghailtean-
stiùiridh – guiding rules):
Bu mhath leam a ràdh gum bi mi a' cumail cho dìleas 's as urrainn dhomhri Riaghailtean-stiùiridh Litreachadh na Gàidhlig (GOC), adh'fhoillsicheadh an 1981 agus a-rithist an 2005, agus gur e sin anfheallsanachd a tha fa-near dhomh anns an leabhar seo cuideachd.(Boyd 2006: 15)
I'd like to say that I keep as faithfully as I can to the 'Guiding Rules ofGaelic Spelling' (GOC), that were published in 1981 and again in 2005,and that will be my philosophy in this book too.
In contrast, there exists the translation bun-tomhas; a compound of bun, 'base,
foundation' and tomhas, 'measurement', which reflects Milroy & Milroy's idea that
the standard is 'a set of abstract norms to which actual usage may conform to a
greater or lesser extent' (1991: 23). Colm Ó Baoill describes his editorial policy in
Duanaire Colach (1997) as following the bun-tomhas:
Tha litreachadh annasach ann an HM, agus san deasachadh seo tha miag iarraidh bun-tomhas na Gàidhlig a leantainn san litreachadh. (ÓBaoill 1997: 49)
There is unusual spelling in HM [the original MS], and in this edition Iwant to follow the standard of Gaelic in the spelling.
The actual choice of the GOC reports, gnàthachas, is rarely used. A semantically
similar choice, however, nòsan, 'customs, manners, habits', is used here by
Seòsamh Watson:
’s ann a rinn mi an litreachadh a leasachadh cho fada is a b’ urrainn acham biodh e ag aontachadh ris na nòsan a tha gan craobh-sgaoileadh an-diugh. (Watson 1997: 33)
What I did was amend the spelling as much as I could till it agreed withthe conventions that are circulated today.
67
Watson's characterisation of nòsan, rather than rules, is in keeping with the
flexible approach he uses to convey the vernacular speech of his informants in
writing.
4.1.1 A Codified Prestige Standard
It is generally accepted that when the orthography was modernised in the mid-
eighteenth century, the variety the orthography was based on was southern, based
in mainland Argyllshire and Perthshire. However, there are few examples of this
variety being chosen due to a belief in its superiority or prestige. Rather, the
writers at the time give prestige to the vernacular, to the general 'Scottish idiom',
in contrast to Classical norms perceived as 'Irish'.
Autonomy from Irish Gaelic
One of the categories of legitimacy that modern language can claim according to
Fishman is autonomy from other languages (Section 2.2). This can lead to Gaelic
writers seeking to distance Scottish from Irish Gaelic. However, it is not so simple
to ascribe this motivation to the frequent comments to removing 'Irishisms' in the
standardisation debate. Often, when using the term 'Irishism', the writer is
referring to a Classical Gaelic form that they wish to replace with a vernacular
Scottish form. It could also reflect the strong anti-Catholic sentiments of the
eighteenth and nineteenth centuries. In the discourse, however, and for whatever
reasons, autonomy from the Irish language is frequently raised in orthographic
discussions of the early nineteenth-century. The first example, from Shaw's
eighteenth-century grammar disagrees with the example of MacFarlane's
translations:
The Rev. Mr. Macfarlane, in his translations and psalms, uniformly usesibh in the dative and ablative plural; which I think resembles too muchthe Irish dialect. Do na muilaichibh, leis na caimbeulaichibh, wouldhave a harsh sound to any provincialist of Scotland. (Shaw 1778: 19)
MacFarlane's translations have been noted as being Classical Gaelic, so it appears
that Shaw uses Irish to refer to outdated forms. Another influential figure, the
Rev. Dr Alexander Irvine who authored a revision of the Bible for the SSPCK and
contributed to the Highland Society's dictionary (1828), was also keen to see
antiquated, 'Irish', spellings removed from the orthography. He wrote to the Rev.
68
Dr John MacLeod who was then in charge of the Highland Society Dictionary telling
him he was too antiquated and 'Irish' in his spelling:
He wished for the Scottish Gaelic language to look presentable to themodern world, and to have an orthography that showed itsindependence from Irish Gaelic. (Hogg 2011: 121)
Irvine's contemporary, the grammarian Rev. Alexander Stewart agreed (see
3.6.2.1 for a brief biography of Stewart). For Stewart and his contemporaries, they
mentioned no 'classical' or shared Gaelic writing tradition, only the inheritance of
'Irish' orthography. In Stewart's account of the vernacularisation and modernisation
of the written language, the underlying narrative was a process of separation from
Irish:
The Scottish writers of Gaelic in general followed the Irish orthography,till after the middle of the last century. However that system may suitthe dialect of Ireland, it certainly is not adapted to the Gaelic of thiscountry. In the Gaelic translation of the New Testament, printed in1767, not only were most of the Irish idioms and inflections, which hadbeen admitted into the Scottish Gaelic writings, rejected, and thelanguage adapted to the dialect of the Scottish Highlands; but theorthography also was adapted to the language. In later publications,the manner of writing the language was gradually assimilated to thatpattern. The Gaelic Version of the Sacred Scriptures lately publishedhas exhibited a model, both of style and orthography, still moreagreeable to the purest Scottish idiom. (Stewart 1801: 39)
Despite Stewart's assessment of the success of the Bible translators, he would later
revise the spelling of the 1820 Pentateuch. The 1826 revision of the Bible,
authorised by the Church of Scotland, found even more 'Irish' to remove:
This edition, although at first intended to be merely a reprint of thatof 1807, with the orthography altered to the system adopted by Dr.Stewart in the revised text of the Pentateuch of 1820, is not strictly areprint, as material alterations were made in expunging, chiefly in theNew Testament, what were considered to be Irish idioms, andsubstituting idioms purely Gaelic. (Reid 1832: 9)
Yet again, when this edition was considered by Father Ewan MacEachen, he too
found even more Irishisms to be expunged:
I saw that the Gaelic Bible, though looked upon as the standard of theGaelic language, had many errors. It has many Irishisms. (MacEachen1842: iii)
69
MacEachen's objections were aimed at the 1801 publication and at the 1826
edition. His contemporary, the poetry editor and lexicographer, John MacKenzie
agreed with his assessment, complaining about:
[...] the Bible, where many words set down in the Irish style are still tobe met with, to the annoyance of intelligent readers and prejudice ofthe Scottish Gaelic (MacKenzie's preface in MacAlpine 1866: x)
MacKenzie praises the Islay schoolmaster Neil MacAlpine's Pronouncing Gaelic
Dictionary (1832) as 'the first to present a Dictionary divested of antiquated
Irishisms' (MacAlpine 1866: x). In his much re-printed collection of Gaelic poetry
Sàr-Obair nam Bàrd Gaelach, MacKenzie himself aimed to 'throw out the Irish'
(MacKenzie 1841: v).
The first half of the nineteenth century is where most of the comments about
'pure' Gaelic and the need for removing Irishisms is found. MacBain is the last to
openly make the same comments:
Celtic scholars, if they find nothing else in the present Dictionary, will,at least, find a nearly pure vocabulary of Scottish Gaelic, purged of themass of Irish words that appear in our larger dictionaries. (...) I guardedespecially against admitting Irish words, with which dictionaries likethose of Shaw and Armstrong swarm. (MacBain 1911: ix-x)
In this section, we have seen that Scottish Gaelic linguists and editors from the
mid-eighteenth to the mid-nineteenth century sought to emphasise the
distinctiveness of Scottish from Irish Gaelic. By doing so, they rejected the
linguistic authority of the previous prestigious variety and based the new standard
on the notion of a 'pure' vernacular Gaelic. The new variety they choose as
'standard' was not the variety of a prestige social elite. Shaw, Irvine and Stewart
were ministers, MacEachen was a priest and they were all interested in spreading
the word of God through an accessible vernacular written norm. The elevation of
the 'purest Scottish idiom' (Stewart 1801: 39), however, led to the new standard
being open to challenge, as the several revisions and criticisms of the Bible show.
In the later decades of the nineteenth century, the linguistic authority of the new
modern Scottish Gaelic standard orthography faced the fragmenting drive of the
'pure' dialects.
70
4.1.2 Codified Conventions
Whether a standard exists or not is often disputed depending on how much a writer
agrees with the existing conventions and how much variation he considers
acceptable. Writers may wish to exaggerate the want of a standard in order to
make their recommendations and reforms seem necessary. The Rev. Dr Alexander
Irvine, before making his own recommendations, is keen to disparage existing
usage:
Those volumes of original beautiful poems, which have, at separatetimes, been offered to the public, are many of them so mutilated, theorthography so imperfect, the sense, in some instances, so obscure,that unless they are revised, corrected, or unfolded, their beautiesmust, ere long, cease to be relished or understood. (Irvine 1801: 12,bold added)
Therefore, his plans for the book Caledonian Bards included the 'settling' of the
orthography.
IV. to demonstrate the capacity of the Gaelic language, and subjoin ascheme for enlarging its bounds, and settling its orthography. (Irvine1801: 4)
This quote also shows how a standard is part of the validation of a language as
within the same point Irvine wants to demonstrate the capacity of the language,
not simply to improve consistency. The preface to the Highland Society Dictionary
also dismisses the early volumes of poetry as following 'no system of orthography':
It is in course of the use and cultivation of languages by writing, that asystem of orthography becomes fixed, and properly conventional. Sucha benefit, has been denied by circumstances to the Scoto-Gaelic; itswritten records being few, and the practice of writing it in latter timeshaving been disused, if we except the few volumes that in recent yearshave, from time to time, been given to the public, of the native poetryand songs; wherein no system of orthography was followed, because thereading or writing of Gaelic was unusual with the compilers. (HSD 1828:xi)
However, the Highland Society Dictionary editors go on to say that the Gaelic
edition of the Bible remedied this situation and provide a standard system:
This deficiency was happily and in a great measure, supplied by thetranslation of the Scriptures, and the publication of them in Scoto-
71
Gaelic, by the Society in Scotland for Propagating Christian Knowledge.The system of orthography followed there, adopted, as it was, bynatives of intelligence and learning in the Scottish Highlands, andimproved by successive editions of the Scriptures, has been strictlyadhered to in the present work. (HSD 1828: xi)
A couple of years later, another dictionary edited by ministers, MacLeod and
Dewar, emphasise the 'justly recognised standard of Gaelic orthography, the
Gaelic Bible':
We consider that one of the excellencies of this Dictionary is an uniformadherence to the justly recognised standard of Gaelic orthography, theGaelic Bible. The venerable translators of the Scriptures, who were socompetent to form an accurate judgment on this subject, gave it themost serious consideration during the many years they were engaged intheir beneficent labours: and feeling as we do, the propriety of entirelyacquiescing in their decision, as well as the advantages which resultfrom uniformity in orthography, we have invariably conformed to therules which they have prescribed. (MacLeod & Dewar 1831: vi)
However, John Mackenzie (1806-1848) ascribes the 'variety of orthographies' used
later in the nineteenth century as partly due to 'the want of a standard' (1877b:
333). The perception of a standard is still relative, however, with other writers
believing that there were 'now almost universally received standards' (Alastair Og
1877: 28). The existence of a standard did not guarantee its dissimulation or
application. By the late nineteenth century, the same complaints about
idiosyncratic spellings are found as, for example, in this paper presented to the
Gaelic Society of Inverness by its treasurer, John MacDonald:
Mar a thubhairt 'Alpein Og', o chionn ghoirid, bha 'n doigh sgrìobhaidhcho caochlaidheach ri breithnachaidhean an luchd-sgrìobhaidh. Machoimeaseas sibh leabhraichean Gàilig a chaidh chuir a mach anInbhirnis, an Glascho, 's an Dùneidin, shaoileadh sibh nach deachriaghailt a chuir sios riamh chum sgrìobhaidh na Gàilige a theagasg, achgu 'n robh gach aon air fhàgail gu saors' a thoil, gu dheanamh mar a bhaceart na shuilean fein. (...) Tha againn fathast ann ar measg beagan dosgoilearan Gàilig, ach tearc mar a ta iad cha chord iad mu 'n doigh a'sfearr air sgrìobhadh na cànaine (MacDonald 1875: 118)
As 'Alpin Og' said recently, the orthography was as variable as thejudgments of the writers. If you compare Gaelic books produced inInverness, Glasgow and Edinburgh, you would think that no rule wasever made to teach Gaelic writing, but that everyone was left to theirown desires to do whatever was right in their own eyes. (...) We stillhave in our midst a few Gaelic scholars, but as rare as they are theydon't agree on the best way to write the language.
72
At the meeting of the Society, members backed MacDonald and the legitimacy of
the Bible as the linguistic authority:
The meeting [of GSI] expressed strong approval of the writer'ssuggestion, that the orthography of the Gaelic edition of the Bibleshould be the basis of Gaelic orthography in general. (MacDonald 1875:121)
In the first half of the nineteenth century, influential writers such as the religious
translators, lexicographers and grammarians hoped to create a standard
orthography in the sense of a set of 'correct' rules.
4.2 The Phonographic System
Many of the world's writing systems are based on a phonographic system where
there are established conventional relationships between graphic and phonic units
(see Coulmas 2003: 89-108). There is also the phonographic ideal which measures
the 'success' of a writing system on how closely it adheres to the ideal (discussed
in 2.1). As the phonographic ideal will always be impossible to reach, there will
inevitably be a compromise with the other factors necessary for standardisation.
Modern Scottish Gaelic orthography is founded on graphic-phonic correspondence
based on the vernacular speech of the writers, grammarians and lexicographers
from the mid-eighteenth to the early nineteenth century. This section will give
examples of how early reformers invoked graphic-phonic correspondence in order
to vernacularise the orthography to meet their literacy goals and how Gaelic's
perceived 'success' measured by the phonographic ideal is used to defend it from
attack.
4.2.1 Phonography and Vernacularisation
From the mid-eighteenth century to the early nineteenth, greater graphic-phonic
correspondence was part of the reforms of the writing system from Classical to
vernacular Scottish Gaelic along with grammatical changes. For example, in Mac
Mhaighstir Alasdair's introduction to his Ais-éiridh na Sean Chánoin Albannaich, he
states his changes are partly to show 'the more usual pronunciation of the
generality of the highlanders' (Mac-Dhonuill 1751: x). When the Rev. Alexander
Stewart outlined his principles of orthography in his introductory section on
73
'Pronunciation and Orthography' to his Elements of Gaelic Grammar (1801), they
were all based on the representation of sound in writing.
Vernacularisation in the form of greater graphic-phonic correspondence was also
seen as an essential step towards making literacy easier to acquire. Thus, in
Alexander McLaurin's First Book for Children in the Gaelic Language (1811) for
example, his rationale for changing <do> to <a dh'> before a vowel is that is makes
Gaelic reading smooth and easy:
Chum leughadh na Gaelic a dheanamh réidh agus furus, tha e féumailgu tugamaid fainear do atharrachadh a ta air a dheanamh air an roimh-fhocal bheag “do,” air an doigh so; an ait’ a radh, do Albainn, do Eirin,do ionnsuidh, is e an gnath a scriobhadh, agus a labhairt mar so, a dh’Albainn, a dh’ Eirin, a dh’ ionnsuidh. (McLaurin 1811: 26)
So that the reading of Gaelic is made smooth and easy, it is useful forus to recognise a change that has been made to the little preposition"do," in this way; instead of saying do Albainn, do Eirin, do ionnsuidh,it is the convention to write, and say like this, a dh’ Albainn, a dh’Eirin, a dh’ ionnsuidh.
When Father Ewen MacEachen advocated reforms in his 1842 dictionary, he had a
mistrust of the 'Irishisms' and 'Englishisms' he saw in existing dictionaries and
Bibles. He took vernacular speech 'from the mouth of the best Gaelic speakers' as
his guide (MacEachen 1842: iv). From this, he used graphic-phonic correspondence
as a guide and justification for his choices:
My principal aim in this Dictionary is to make the spelling echo with thesound in all cases where uniformity is wanting (MacEachen 1842: vi-vii)
However, MacEachen's guide to which sound should be used was firmly based on
'my own dialect, with its proper and uniform orthography' (MacEachen 1842: iv).
This illustrates how the drive for vernacularism could also result in the promotion
of dialects which then competed for acceptance within standardisation, as will
be discussed in the following sections.
4.2.2 The Phonographic Ideal in Scottish Gaelic
The phonographic ideal is one of the main ways that orthographies are judged to
be successful; evaluating an orthography according to how well it meets the ideal
of one grapheme, one sound (see Section 2.1.3). This section will consider the role
74
of the phonographic ideal in Scottish Gaelic. In the historiography of Scottish
Gaelic orthography, there is a complex relationship between recognising the
impossibility of the phonographic ideal and appealing to it to defend the
language's status.
The limitations of the phonographic ideal are discussed by the Rev. Alexander
Stewart in his grammar when he asks:
How near ought the written language to correspond to the spoken; andwhere may a disagreement between them be allowed with propriety?(Stewart 1801: 26)
Stewart makes significant effort to discuss the factors involved with his readers.
He is likely to have expected his readers to also be scholars and educated people.
He uses his introduction to persuade his readers of the necessity for continuing to
modernise the orthography and engages with the theoretical questions of what
spelling in an alphabetic system is trying to do: balancing the codification of
speech with the impossibility of following all the changes and variabilities of
speech particularly as he is aware that there is:
a diversity of pronunciation (...) in different districts of the Highlandsof Scotland, even in uttering the same words written in the samemanner (Stewart 1801: 2).
He concludes that the 'disagreement' between letter and sounds is necessary for
the sake of 'preserving some degree of uniformity' (1801: 29).
The authors of the Highland Society dictionary similarly noted that a perfect
phonographic ideal is not possible although a 'regular system', in other words a
codified relationship between sound and grapheme, allows for writing to function
as graphic communication:
though it be not possible to represent sound, by any notation of letters,with sufficient accuracy and plainness; yet the greatest facilityattainable, in the absence of oral communication, is afforded by aregular system of such notation. (HSD 1828: xi)
MacAulay argues against the phonographic ideal, recognising that the relationship
between the sound and the representation of it in writing should not be overstated
75
or even aimed for: ‘It is essential therefore not to confuse speech and writing and
not to act as if they were the same’ (1976-78: 82):
Even if it were true that an orthographic system were to begin as adirect representation of the spoken word (i.e. be something like a broadphonetic transcription) since it is a system with conventionalcharacteristics, which are necessarily conservative in nature in order tomaintain the integrity of the system, and since a basic feature of spokenlanguage is that it changes, then, before long, the directlyrepresentational nature of the system must change (MacAulay 1976-78:83).
It appears then, that the phonographic ideal has been recognised as being
impossible to achieve; that graphic-phonic correspondence will be bound by the
requirements for uniformity and regularity which are necessary for functional
written communication. However, this has not meant it has not remained as a
measure of success with which to compare the Scottish Gaelic writing system to
others.
4.2.3 Phonography as Gaelic Superiority
As was noted in 2.2.1, standardisation is a key element in European societies'
understanding of what makes a certain language variety a valid language and the
phonographic ideal is one of the ways in which a language standard is measured.
Therefore, relative success in terms of the phonographic ideal can be used to
praise and defend a language from attacks on its legitimacy. The phonographic
ideal is often invoked to praise the Scottish Gaelic system. Writers and
commentators argue for the superiority of regular graphic-phonic correspondence
in Gaelic as compared to others, in particular, as compared to English spelling. In
this way, they defend Gaelic from being attacked as nonsensical (see also Section
4.5).
Shaw is the first to elucidate this. He argues that, according to its internal
orthographic conventions, Gaelic has more regular and consistent correlation
between spelling and pronunciation than Irish, English or French:14
14 Conversely, after Irish Gaelic underwent significant reform in the twentieth century, its greaterphonographic correspondence could be used to show its superiority to Scottish Gaelic as in thisexample: 'Le Gàidhlig "standard" air a teagasg anns na sgoiltean air fad, agus litreachadh ùrsimplidh (nas fhasa na an dòigh-litreachadh uabhasach duilich a tha againn ann an Alba co-
76
Unlike the Irish, the Scotch Galic delights to pronounce every letter,and is not bridled over with so many useless and quiescent consonants.The English and French are infinitely more difficult to read andpronounce and have many more silent and mute letters. In the Galicthere are no such ugly looking words as thought, through, strength, &c.nor sounds so different from what the letters at other times express.(Shaw 1778: xx-xxi)
Although Malcolm MacFarlane created his own phonographic system, he still
praised the 'general consistency' of the existing writing system and pointed out its
superiority over English:
The orthography of the Gaelic language does not consistentlycorrespond to the spoken sounds. [...] Eighteen letters are made tofulfil more than double that number of functions. Phonetic evolution is,to a certain extent, recorded graphically. Yet, notwithstanding thesecircumstances, such is the general consistency of the system that it iseasier to learn the reading of Gaelic than that of English. (MacFarlane1889: 67)
At the start of the twentieth century the editor of the Celtic Annual also noted
that the success of Scottish Gaelic in meeting the phonographic ideal places it
ahead of other languages:
Phonetically, the spelling of our Scottish Gaelic is far in advance of IrishGaelic, and very far indeed in advance of English. In fact it may be saidthat the men under whose hands our spelling system developed knewsome phonetic facts which have hardly dawned on the ordinary Englishphonetician's mind, and gave effect to them in the spelling of ourlanguage. And yet some of those English phoneticians have theimpudence to say, before knowing anything of our language's spellingsystem, that it is the most unphonetic of all. (MacLeod 1913: 8)
This then forms a reinforcing pattern: the better the adherence to graphic-phonic
correspondence, the more 'worthy' the orthography (and by implication, the
language) is considered to be; and the more adherence to this ideal is considered
the measure of a good orthography, the more inconsistencies and exceptions need
to be 'ironed out' or 'tidied up' so as not to threaten the status that the ideal
confers. Thus, the change from tigh to taigh, so to seo, to adhere to the
conventional graphic-phonic correspondence. In another example, Stewart terms
dhiù) [With 'Standard' Gaelic taught in all the schools, and a new simple spelling (easier anywaythan the terribly difficult spelling we have in Scotland] (MacAoidh 1978: 74)
77
as 'defects' elements that do not adhere to the phonographic ideal, as in this
example regarding the unmarked mutation of <l>, <n> and <r>:
both classes of consonants are alike mutable in their pronunciation; andtheir mutation ought to have been marked in the orthography, thoughit has not. This defect in Gaelic orthography has been often observedand regretted, though it has never been corrected. (Stewart 1812: 3)
When recent writers describe spelling changes as sgioblachadh, 'tidying', what
they are usually referring to is greater adherence to a graphic-phonic
correspondence. It is the adherence to this principle that usually wins the battle
of ideas and motivates nearly all of the changes and adjustments that shape Gaelic
orthography through the nineteenth and twentieth centuries.
4.2.4 Phonography, Dialects and Uniformity
The basis of the graphic-phonic correspondence was on the phonological features
of southern Gaelic dialects, particularly Argyll and Perthshire. However, as part
of vernacularisation, dialects were given prestige as the 'true pronunciation' in
contrast to the conventions that Classical Gaelic encoded, as in this quote from
Irvine:
It will endeavor to ascertain the leading characteristics of the provincialdialects, the true pronunciation, of the Gaelic Language. (Irvine 1801:6)
As the same time as giving the dialects prestige, Irvine is conscious of the need to
know what they have in common, in the form of their 'leading characteristics'.
That Gaelic writers have continually needed to find a balance is illustrated by
Archibald Sinclair. Prefacing An t-Oranaiche [The Songster] (1879), Sinclair writes:
dh’fheuch mi ris a’ Ghàilig a litreachadh mar a tha i air a fuaimneachadhann an ceàrnaibh air leth. So puing a tha duilich a laimhseachadh gunceòlmhoireachd agus co-sheirm cuid do na h-òrain a mhilleadh, (…) agusged nach biodh gach ni a réir meidhean-tomhais nan tiolpairean, bumhaith leam gu ’n cuimhneachadh iad an sean-fhacal—“Saoilidh am feara bhios ’n a thàmh gur e féin is feàrr làmh air an stiùir.” (Mac-na-Ceàrdadh 1879: vii)
I tried to spell the Gaelic as it is sounded in separate districts. This isa difficult point to manage without ruining the musicality and harmonyof some of the songs, (…) although not everything might comply with
78
the yardstick of the grumblers, I'd like them to remember the proverb– "The man who's at rest thinks he is the best at the helm".
This is discussed further in section 4.3 and in the study of the use of <eu> and <ia>
in Chapter 6.
4.2.5 The Compromises of Phonography
Phonographic correspondence is generally understood to be an impossible ideal
and several writers discuss how it is limited. Instead, they agree that Gaelic has a
phonographic basis while recognising that it will be a mixed system; compromised
by the inherent differences between speech and writing, the continuity of
retaining historical features, the avoidance of homographs and the compromise
between dialects. The notion of phonographic ideal as a measure by which writing
systems are judged, however, is invoked on occasions when the writer is seeking
to defend the legitimacy of Gaelic. There has, over the decades, been a shift to
greater graphic-phonic correspondence. It is used to remove exceptions from the
conventions of graphic-phonic correspondence but is set aside when reducing
homonyms.
4.3 Dialects and Acceptable Variation
Dialect usage can threaten the status of written standard as variation and
uniformity is seen to weaken it. This in turn can threaten the validity of the
language itself. Yet, dialects in Gaelic carry prestige as authentic forms of the
language. This stems from the eighteenth century spelling reformers. They
appealed to vernacular language use as an authority to counter the prestige
Classical Gaelic standard as opposed to prestigious authors. This was followed in
the mid-nineteenth century by the interests in folklore spurred by John F.
Campbell's Popular Tales of the West Highlands. Once again, authority and
authenticity came from vernacular sources and this was reflected in orthographic
choices to reproduce tales as they were told. In this section, examples where
variation in orthography has been both discouraged and encouraged are discussed.
79
4.3.1 The Curbing of Dialects
One of the ways in which a standard is maintained is the discouragement or
prohibition of dialects. Non-standard usages in Gaelic are disparagingly called
localisms or provincialisms as a relatively mild way of discouraging their usage.
There are also examples of writers explicitly warning of the fragmentation that
dialectal writing threatens. In the arguments between the Rev. Dr Alexander Irvine
and the Rev. Dr Thomas Ross over the Bible translations, Irvine warns of
consequences of dialectal variation:
Were this overture listened to, what would be the consequence? Why,this, that as in the county of Ross various dialects prevail, and as thistranslation of Dr Ross cannot be adapted to them all, the samecomplaint would never cease to exist. (Ross 1821: 4)
A similar point is made by MacLeod in 1913 where he sees too much dialectal
variation at work and 'confusion and delay' with the lack of a common standard:
At present nearly every person who attempts Gaelic work of a kind,which in other languages would be confined strictly to the literary style,uses his or her own provincial or local dialect. Provincial and localdialects have their values and uses, and inside their own sphere theyought to be recognised; but nothing but confusion and delay can resultif there be not a literary dialect as in all other languages, forming acentral standard round which all can rally, no matter what theircolloquial style of speech may be. (MacLeod 1913: 7)
Dialectal variation could be a chance for humour as in Eoghan a' Chaolais' replyto another contributor's lofty ambitions in An Teachdaire Ùr Gàidhealach (1835-). In a sarcastic joke, he blames the Ross-shire folk for pushing spelling reform.
Tha "Mactalla" 'g ìnnseadh gu'n d' ionnsuich e iomad òran obhuachaillean Thyre. Tha mi 'deanamh dhe gur e so cuid de'n speileadhùr tha muinntir shiorramachd Rois a' feuchainn a chur oirnn, oir 's ann ab' àbhaist duinne a sgrìobhadh, Tirie (Eoghan a' Chaolais 1835: 43).15
'MacTalla' tells us that he learnt many songs from the shepherds of Tyre[Greek philosophers]. I guess this must be some of the new spellingthat the Ross-shire folks are giving us, because we used to write Tiree.
This exemplifies a theme where unfamiliar or unapproved of forms in Gaelic are
dismissed as 'those other folks' weird dialects'. Another example is in the account
15 'MacTalla' was a pen-name of Lachlan MacLean (1798-1848) from Coll, the editor of AnTeachdaire Ùr Gàidhealach (1835-36).
80
by the Rev. Kenneth MacLeod (1871-1955) of Prof. John Stewart Blackie's visit to
Eigg in his youth:
Bha e mòran na b’fheàrr air a’ Ghrèigis. Gun teagamh, leughadh e agusdh’eadar-theagaicheadh e Gàidhlig, ach cha deachaidh aige riamh air abruidhinn air a leithid a dhòigh is gun tuigeadh neach sam bith dè bhae ciallachadh. Co-dhiù, cha do thuig sinne aon fhacal a thubhairt e. Achbha na h-Eigich cho modhail ’s nach aidicheadh a h-aon diubh sin. “Nacheil fhios gu bheil Gàidhlig mhaith aig an duine,” ars iadsan, “ged nacheil sinne ga tuigsinn. ’S fheudar gur h-ann fada mu dheas a dh’ionnsaiche i.” Bha tè anns an eilean a thug bliadhna no dhà ga cosnadh fhèin annan Loch Raonasa an Arainn, agus an uair a chuala i mu dheighinnGàidhlig Mhic’Ille-dhuibh, “Cuiridh mise geall,” ars ise, “gur h-iGàidhlig Arainn a th’aig an duine agus mas i, cha bhi i idir furasda atuigsinn.”
He was much better at Greek. Without doubt, he could read andtranslate Gaelic, but he never managed to speak in a way that anyonewould understand what he was saying. Anyway, we didn't understand asingle word he said. But the Eigg folk were so polite that none of themwould admit that we didn't understand him. "The man must have goodGaelic', they said, "even though we don't get it. He must have learnt itway to the south." There was one women on the island that had spenta year or two working in Lochranza on Arran, and when she heard aboutBlackie's Gaelic she said, "I bet that he has Arran Gaelic and if he does,it wouldn't be easy to understand". (MacLeod 1988: 64-5)
When John MacKenzie compiled An English-Gaelic Dictionary to accompany Neil
MacAlpine's Gaelic-English Pronouncing Dictionary (1832), he first praised
MacAlpine's orthography for being 'much more correct' that those who preceded
him (MacAlpine 1866: vii). He then goes on to criticise MacAlpine for his 'various
provincialisms' which MacKenzie believes have no place in a dictionary.16 Archibald
Farquharson, from Perthshire, singled out Lewis Gaelic in particular as 'provincial':
As there is a good deal of provincialism in the Gaelic of Lewes, I think,in writing, it would be better if possible to follow our standard ofGaelic. (Farquharson 1868: 15)
In spite of these concerns, a counter-narrative supported the use of dialects,
particularly in the contexts of poetry (or song) and folklore.
16 ‘These he defends, giving them a preference over words of more general acceptation, aproceeding by no means to be justified in a work of this kind’ (MacAlpine 1866: vii)
81
4.3.2 Poetry and Folklore
In the quote used in the previous section by MacLeod, he argued for a common
standard yet said that 'provincial and local dialects have their values and uses,
and inside their own sphere they ought to be recognised' (MacLeod 1913: 7).
Donald MacKinnon sets out that these 'spheres' when 'localisms' are acceptable
are: in poetry where 'the ring of the line must be preserved' and 'all local sounds
that lend grace and melody to his rhymes'; and in texts that are 'illustrating
dialect, or registering dialectal material for linguistic and historical purposes'
(1909: 17). This paradigm, although most clearly codified by MacKinnon, is
commonly accepted by other writers. Given Nicolson's interest in proverbs, it is
not surprising that he prefaces his Collection of Gaelic Proverbs and Familiar
Phrases (1881) by arguing in favour of variability and flexibility:
We have as yet no absolute standard of Gaelic orthography, and it is nodisgrace considering that William Shakespeare spelled his own greatname in several ways, and that even Samuel Johnson's English spellingsare not all followed now. (Nicolson 1881: x)
The folklorist Alexander Carmichael (1832–1912) discussed the influence of John
F. Campbell of Islay on his methodology. He notes, as Campbell did before him,
that spelling variations may be used in the text to show variation in speech:
Some localisms are given for the sake of Gaelic scholars. Hence thesame word may be spelt in different ways through the influence ofassonance and other characteristics of Gaelic compositions.(Carmichael 1900: xxxi)
His daughter Ella Carmichael also noted when editing songs 'I have retained
'localisms' which seem in keeping with folk-songs' (Carmichael 1904: 147). It is
possible, although more rare, to find justifications of dialectal variation extending
into other areas of literature. However, Kenneth D. MacDonald, the editor of the
short story collection Briseadh na Cloiche (1970), which is included in the corpus,
said:
Ann a bhith ’gan deasachadh air son a’ chlò, rinneadh beagansgioblachaidh air litreachadh an siod ’s an seo, ach dh’fhiachadh ri dual-chainnt gach sgrìobhaiche a ghleidheadh mar a bha i. (MacDonald 1970:9)
82
In editing them for printing, a little tidying of spelling here and therewas done, but an attempt was made to keep each writer's dialect as itwas.
MacKinnon's paradigm is explored further in Chapter 6 where it is an important
factor in the variability of the use of <eu> and <ia>.
4.3.3 Unacceptable Variation
Dialectal acceptability does not extend to all spoken forms and forms seen as
colloquial are often rejected. MacKinnon acknowledges that compared with plain
prose 'conversation will also insist on having its privileges.' However, he does not
believe that the written word should closely follow conversational forms:
But we have allowed the colloquial to preponderate in our writing; andas a consequence the Apostrophe runs riot on a Gaelic page. Why shouldwe persist in writing 'Se for Is e, 'it is'? Everybody says, but why shouldany one write, Co 'th' ann? for Co a tha ann? 'Who is there'? Our fathersand grandfathers were in these respects more careful and accurate thanwe have become. (MacKinnon 1910: 197)
William J. Watson, the place-name scholar, also agreed:
Contracted forms of words - the bane of modern written Gaelic - havebeen freely extended (...) At present there is a tendency to make thewritten language reproduce all possible contractions of commonspeech, and the result is unsightly and often obscure. The soundprinciple is to write each word in full, except in the case of recognisedcontractions of old standing. (Watson 1915: vii)
Donald MacAulay was also not impressed with 'aggressively colloquial' spelling:
there has arisen a tendency with some people to use colloquial languagefor all levels of discourse, and furthermore to spell it in an aggressivelycolloquial fashion. (1976-78: 88)
Spelling variation can be acceptable, in Gaelic, but only in its place. Variation is
not considered acceptable if it perceived as ad hoc, or based on colloquial speech.
The polynomic model was discussed in section 2.2. It was suggested that, as in
Guernsey, the Gaelic community has a folk linguistic acceptance 'of regional
variation as a source of richness' (Sallabank 2010: 311). This section has shown
that this is not simply 'folk linguistic' in the Gaelic context but that this extends
83
to (or possibly derives from) scholarly works and literary authors. The early
religious codifiers valorised vernacular speech and the dialects of Scotland in
contrast to Classical norms regarded as Irish. From the mid-nineteenth century
dialects were valorised by folklorists and historical linguists interested in 'pure'
language. We have also seen that there is a tension between the prestige of
dialects and the belief in the necessity of a common written norm.
4.4 Etymology and Historicism
Etymology can be important in language ideology as it stems from what Fishman
calls 'historicity' or the way in which history is considered to socially validate a
language (Section 2.2.1). In terms of orthography, this belief suggests that spelling
should preserve etymological information. This can be considered a worthy goal
in and of itself or to assist readers with their knowledge of older forms of the
language. Etymology therefore can be invoked when it comes to arguing for or
against a particular form. Consistency with the past or spelling conservatism is
related to this, but not equivalent. The etymological principle is interested in the
oldest form, conservatism with preserving the status quo even if it is
etymologically inaccurate. This section will discuss where etymology is cited as
orthographic principle, how it clashes with the phonographic ideal and how it
relates to historicism and spelling conservatism.
Historicity is visible, particularly in the early texts which have to validate the
language. The idea of providing an ancient lineage is important in order to
promote the validity of the language as not simply 'a barbarous tongue'. For
example, Shaw spends a good deal of time in his introduction to his Analysis
describing Gaelic as the descendent of the 'Celtic nation' which was marginalised
in preference to Greece and Rome. He stresses the antiquity of Gaelic, introducing
it by talking of its history at the time of the Roman empire, equating Gaelic and
Celtic history with the prestigious great civilizations (Shaw 1778: ix-xi).
4.4.1 Etymological Principle
At the end of the eighteenth century when Shaw was creating his grammar, he
references advice that appears to promote the phonographic ideal and the idea
of reforming the writing system:
84
Being the first that has offered the public a grammatical account of theGalic, it was recommended by several persons to frame a new alphabet,consisting of letters or combinations, to express all the sounds in thelanguage, without any mute letter. This is impracticable; but though itcould be effected, it would only render the etymology more perplexing.(Shaw 1778: xx)
Shaw disagrees with the idea by appealing to etymology. Stewart can also appeal
to etymology as a deciding factor, as in this example of the spelling of 'Gaelic'
itself:
The word 'Gaelic' has of late been written with ae in the first syllable.Whether this way of spelling it be preferable to the former 'Gailic,Gaidhlig, Gaoidheilg, &c.' must depend on the etymon of the word; apoint not yet fully settled. (Stewart 1801: 5)
Alexander MacBain's etymological interests are obvious from his Etymological
Dictionary (1896). This meant that although MacBain praised MacEachen's
lexicography, he found it etymologically wanting and emended 'on a large scale'
MacEachen's orthography in his 1902 jointly edited edition (MacEachen 1979: vi).
MacBain and Whyte's editorial comments show their preference for etymology:
'While giving a common, though historically wrong, spelling of a word, they make
cross-reference to its proper form' (MacEachen 1902: vi).
In 1933, John Lorne Campbell published a short article on standardising Gaelic
spelling in An Gaidheal. In the article, his interest is clearly prescriptive, being
more interested in 'correcting' contemporary norms rather than standardising as
uniformisation. Campbell stresses the importance of etymological origin:
Similarly there are words now etymologically wrongly spelt in ScottishGaelic in an attempt to reproduce their pronunciation morephonetically. Of these the word aobhar, correctly adhbhar is a goodexample. Because the pronunciation of its first syllable resembles insome dialects the sound written ao in taobh etc., its correct spellinghas been abandoned for the “phonetic” one. (...) The best thing to dowith this kind of word is to leave the original spelling alone or at leastnot alter it to something fundamentally different from its etymologicalorigin. There are a number of words in Gaelic which have suffered inthis way, as Dr Alexander Cameron (Reliquiae Celticae) and others havepointed out. Any revision of Gaelic spelling should insure theircorrection. (Campbell 1933: 26-27)
85
Campbell in this quote clearly considered the etymological spelling to be the
'correct' one in contrast to simpler graphic-phonic correspondence.17 Campbell
also argued for other etymological spellings, dismissing the <eu> forms that were
in use for over a hundred years before he wrote as 'wrong':
thus writing correctly céim, léim, féim, for ceum, etc., which are wrongspellings, the m in these words being originally a narrow consonant; andbréag, léag, etc. for breug, etc. This would have the merit of gettingrid of an unnecessary digraph, and also bringing Scottish Gaelic into linewith Irish, which has preserved the correct form. (Campbell 1933: 26)
Campbell’s recommendations for regularising unstressed vowels and regularising
<ao> in this article were, eventually, realised with GOC. His option for replacing
<eu> with <éa> and <éi> before broad and slender consonants respectively was
not taken up. The former two advocated regularising to one option from those in
use while the latter recommended adopting etymological Classical forms no longer
in conventional usage. This demonstrates a clash between conservative forms
(those in use), and etymological forms no longer in use where the conservative
form wins out.
In a more recent example, in Richard Cox's review of GOC 2009 (Cox 2010), he
evaluates the recommendations in part by assessing how they relate to Old Gaelic,
e.g. 'The form siad ‘hero’ (Word List) fails to acknowledge the word’s etymology
(OG sét, cf. OG fér > modern feur)' (Cox 2010: 165).
4.4.2 Etymology and Compromise
Etymological goals often conflict with the phonographic ideal and phonographic
correspondence. Etymology can also be in conflict with the demands of
standardisation for uniformity. The conflict between phonography and etymology
is summed up by the Rev. MacLean Sinclair in his 'Speculations on Orthography':
We should spell words, so far as practicable, just as they arepronounced. [...] We should preserve the oldest form of words, so faras that can be done without violation to the present mode ofpronouncing them. (MacLean Sinclair 1898: 271)
17 The example Campbell uses, adhbhar not aobhar, is the current recommended form (GOC).
86
In this brief account, MacLean Sinclair outlines two competing principles. He
balances them by hedging that they should be applied 'so far as practicable' and
'so far as that can be done'. In this, he epitomises the compromise of real writing.
Conservatism in spelling can also conflict with etymology, as the example from
Campbell showed earlier. The conservative impulse is part of standardisation's
requirements for uniformity: in this case consistency over time. The focus is not
on etymology, but a resistance to a change to conventions. The theory that this
has existed in Scottish Gaelic writing has been espoused by scholars such as Eric
P. Hamp, who comments in an introductory section to the Linguistic Survey of
Gaelic Dialects that conservatism has a 'strong sense of observance':
The distinction between language and dialect for Scottish Gaelic is tobe mapped largely on an inherited literary and bardic norm; while thereis no national or academy-promulgated or school-imposed set of rules[…], there has existed a strong sense of observance before the practicesof preachers and learned persons. (Hamp in Ó Dochartaigh 1997: 8)
Conservatism means that writers and editors can be reluctant to innovate. In this
example from Katherine Whyte Grant she cites the needs of those acquiring
literacy, children in this case, for conservatism:
Tha mi air cumail ris an t-sean dòigh-sgrìobhaidh. Tha clann chocleachdta ris, ’an leughadh a’ Bhìobuill, agus leabhraichean eile, gu ’ndo mheas mi nach biodh e glic atharrachadh a dheanamh, no dòigheanúra a ghnàthachadh. (Whyte Grant 1911: 4)
I have kept to the old spelling. Children are so used to it, reading theBible, and other books, that I judged that it would not be wise to makechanges, or utilise new ways.
Donald MacKinnon also admits, while complaining that previous scholars chose to
write ar n-athair and gu'n iarr, instead of either arn athair and gun iarr, or ar n-
athair and gu n-iarr, that 'It would probably serve no useful purpose for writers of
Scottish Gaelic to change their practice now' (MacKinnon 1905: 11).
4.4.3 Summary
In this section, we have seen that there have been few cases of etymology being
used to inform spelling choices over the modern period. It is conservatism, rather,
that seeks to preserve familiar spellings but not reinstate older norms or decide
87
which is 'correct'. When it comes to reform, etymologically 'correct' or
conservative forms often lose out to phonography. This is not always the case,
however. As will be seen in Chapter 6, conservatism wins out with <eu> being
preferred to <ia>. The conservative form of <sd> in many lemmas loses out to
standardisation's demands for uniformity and simplicity to <st> in all lemmas in
Chapter 7.
4.5 The Fear of Illegitimacy
In Chapter 2, it was noted that without standardisation, one of the pillars of
cultural legitimacy that leads people to consider a language variety a 'real'
language is removed. If it has no standard (i.e. uniform) writing system, it is not
a legitimate language. To counter the attacks, Gaelic writers also draw on
ideologies of spelling as defence. In this section we will see examples of spelling
being used to attack Gaelic, examples of where Gaelic writers exhibit the fear of
linguistic illegitimacy and where they defend Gaelic orthography. There are three
particular narratives that writers use when defending Gaelic spelling. The first is
the use of the phonographic ideal to rate Gaelic more highly than other languages,
discussed earlier in this chapter. The second is to appeal to prestige languages –
Classical or other European majority languages – to argue that a level of variation
is normal. The third is the narrative that Gaelic spelling is simple, effective and
sufficient.
Over the centuries the legitimacy of Gaelic has been repeatedly challenged. An
early example of the perception of Scottish Gaelic orthography is found in Samuel
Johnson's comment that:
Whoever therefore now writes in this language, spells according to hisown perception of the sound, and his own idea of the power of theletters (Johnson 1775: 267-68)
Johnson was followed by other writers such as John Pinkerton who said in his
Enquiry into the History of Scotland (1789) that:
of all etymology whatever, the Celtic is the most uncertain, becausethe language is hardly a written one, and its orthography, on whichetymology depends, is quite various and lax. (Pinkerton 1789: 139)
88
The Rev. Dr Alexander Stewart begins the introduction to his grammar by
acknowledging that the work will be 'variously appreciated' particularly by those
who 'will be disposed to deride the vain endeavour to restore vigour to a decaying
superannuated language' (1801: 1). Stewart responded by criticising those 'who
judge of the language only from its appearance in writing [who] have taken
occasion to vilify it as ''unfixed and nonsensical''' (1812: 38). These opinions also
led the Rev. Dr Irvine to complain that,
Strangers to the Gaelic language have taxed it with unwarrantableharshness. Deceived, or misled, by a dress, with which they were notacquainted, they passed too hasty a judgment. (Stewart 1801: 10)
The typical tropes of denigration are that Gaelic has no spelling conventions at all
(as with Johnson above) or that Gaelic has no uniform standard (it is too various
in its spelling). Sometimes, there is more general prejudice such as this by an
unknown reviewer in Nature of the newly-created Scottish Celtic Review:
The programme is an excellent one, (...) as well as to afford room forthe discussion of questions relating to Gaelic grammar and orthography.This last, it seems to us, is a subject with which the Gaelic scholars ofthe Highlands trouble themselves a great deal too much. Modern Gaelicorthography, whether in Ireland or in Alban, is simply incorrigible, andhad better be left alone for the rest of the natural lives of the survivingdialects. This involves no great inconvenience; for no scholar who wantsto understand the history of a Gaelic word ever thinks of being guidedby any of the modern spellings which may be in use, but goes back tothe Irish of the Middle Ages, or farther still, to what is technically knownas Old Irish. It is some consolation to Englishmen to know that Englishorthography is not quite the worst in the world, and that Tonald seldomwrites, but that when he does he spells more outrageously than themost wayward spelling-book ever known in the land of the Southron.(Anon 1881: 50-1)
Donald was a stock name in the stereotypical portrayal of Highlanders in the
nineteenth century and often spelled as 'Tonald' to mock the Gaelic pronunciation.
Even at the beginning of Celtic lexicography, Lhuyd needed to defend variation in
spelling:
This Uniting of some words, and the Disjoyning of others, as also theVariation of spelling, occur also in the old Latin manuscripts, as well asthose of other Languages. (Lhuyd 1707: 232)
89
Lhuyd normalises the idea of spelling variation with his readers by appealing to
the high prestige of Latin. At the end of the eighteenth century, Shaw also
anticipated that his readers would expect a fixed standard. He argued that its
relative lack of printing history was the cause of continuing variation:
A speech so singular in its inflections, so ancient in its structure, and socopious in words as the Gaelic, although much written, cannot be sofixed in its orthography as those tongues which enjoy the benefits ofPrinting. With the Irish, however, (whose dialect has always been thewritten and studied language) the difference was very inconsiderable.One vowel or diphthong is sometimes substituted for a similar one, andthe commutable consonants and combinations are interchanged. Thishas been the case with the Greek and Latin languages, as well as withthe Gaelic. (Shaw 1780: 8)
Just as Shaw also appeals in the previous quote to the Classical languages of Greek
and Latin for legitimacy, Stewart appeals to the universality of the processes of
change which means that Gaelic 'has shared the common fate of all written
languages' (1812: 24). By comparing Scottish Gaelic to Greek, Latin, English and
French orthography he seeks to demonstrate that these are not unique or strange
aspects of Gaelic orthography (1812: 26). Thus Stewart explains broad and slender
consonants by invoking examples from English and Italian:
It may not appear obvious at first sight, how a vowel should beemployed, not to represent a vocal sound, but to modify anarticulation. Yet examples are to be found in modern languages. Thusin the English words, 'George, 'sergeant,' the e has no other effect thanto give g its soft sound; and in 'guest, guide,' the u only serves to give gits hard sound. So in the Italian words 'giorno, giusto,' and many others,the i only qualifies the sound of the preceding consonant. The same useof the vowels will be seen to take place frequently in Gaelicorthography. (Stewart 1812: 2)
Despite Dr Johnson's controversial pronouncements on Gaelic in his A Journey to
the Western Isles of Scotland, MacLeod and Dewar made subsequent reference to
his lexicographic authority in order to justify 'anomalies':
'Every language," says the great lexicographer, Dr. Johnson, "has itsanomalies, which, though inconvenient, and in themselves onceunnecessary, must be tolerated among the imperfections of humanthings, and which require only to be registered that they may not beincreased, and ascertained that they may not be confounded.' (MacLeod& Dewar 1831: vi)
90
The anonymous writer 'Golisdar Gàelach' more aggressively attacked variation in
English spelling in his letter to the editor of the Scots' Times:
The feigned beauties of the Anglo Saxon or Johnsonian English, now thefavourite dialect, and most fashionable in the un-it-ar-i-an kingdom ofGreat Britain and Ireland, are given in a rhyme of rules, which are allsmothered and crushed to death, or left in the minority by a hugemultitude of exceptions every where, so that wholesome rules areindeed very much wanted. (Gàelach 1830: 1)
By referencing other languages with social prestige and by citing those considered
to have linguistic authority, Gaelic writers hoped to persuade others that some
variation was not a sign of linguistic barbarity.
The other defence narrative is to appeal to the 'virtues' of simplicity, effectiveness
and sufficiency (i.e. in that it has no need for new letters). This is summarised by
John Whyte and Alexander MacBain who reassure readers of their dictionary that
Gaelic orthography:
[...] though it appears somewhat complex to the eye, is at once simple,effective, and quite sufficient for all the orthographical requirementsof the language. (Whyte & MacBain 1902: 1)
James Munro, in the orthographic description of his Practical Grammar, praises
Gaelic effectiveness in showing lenition; arguing that retaining the primary letter
and adding an <h> is superior to the Welsh system 'in point of perspicuity' (1843:
2). One of the founders of the Gaelic Society of Inverness implored members to
improve their literacy in Gaelic and tried to reassure them of the orthography's
simplicity:
It is generally taken for granted that Gaelic is a language difficult tolearn – impossible to spell; but had those who are of this opiniondevoted to Gaelic but the hundredth part of the time which theydevoted to English before they became masters of its orthography, theirtale would be very different. The principles of Gaelic orthography arefew and simple; and if these principles are mastered, the language ismastered. If you speak Gaelic, I assure you that two months' earneststudy – two hours every evening – will result in you being able to writeit. (MacKay 1873: 44)
John MacKenzie, in his 'Defence of the Orthography of the Gaelic Language' (1877),
offers accounts of people mastering the system in a few months to show its
91
simplicity and effectiveness (1877b: 334). He references a report of the Glasgow
Auxiliary Gaelic School Society where is it noted:
that to acquire the art of reading Gaelic to one who speaks it, is by nomeans so formidable a task as it is for one who speaks English to learnto read English. (MacKenzie 1877: 334)
Another level of meta-linguistic analysis emerged in the late twentieth century,
as concerns were raised that the continued discussions around spelling, rather
than just variation itself, reflected badly on the status of language:
Tha an stràc geur fhathast beò – air èiginn. Ach am bu chòir a bhith?Mura h-eil na Gaidheil comasach air tighinn gu aonta air rud chobunaiteach, carson a bhiodh an saoghal mòr a-muigh a' toirt spèisdhaibh is don cànan? (MacIlleathain 2005)
The acute accent is still alive – just. But should it be? If the Gaels aren'table to come to an agreement on something so basic, why would thegreat outside world give us and the language respect?
However, if there were no debates on spelling at all, it would, in fact, be quite
unlike the great outside world given that Germany, France, Spain, the Czech
Republic and many others have undergone and debated reform in the twentieth
and twenty-first centuries.
4.6 Summary and Conclusions
The battle of ideas and ideologies that surround discussions of orthography means
that any set of conventions is a result of compromise. It also means that shifting
language use or language change engenders further discussion. This discussion
itself has often been criticised. In a paper read to the Gaelic Society of Inverness
entitled 'Am Feum a ta air aon doigh suidhicht' Aithnichte air Sgriobhadh na
Gaidhlige' (The Need for One Settled Recognised Way for Writing Gaelic), a Mr
John MacDonald expresses his dislike of writers that do not follow the example set
by the Bible. He does not see Gaelic scholars setting a good example:
Tha againn fathast ann ar measg beagan do sgoilearan Gàilig, ach tearcmar a ta iad cha chord iad mu 'n doigh a's fearr air sgrìobhadh na cànaine(MacDonald 1875: 118)
92
We still have in our midst a few Gaelic scholars, but as rare as they arethey don't agree on the best way to write the language.
When Dr Hugh Cameron Gillies was Secretary to the Gaelic Society of London, he
proposed a resolution aimed at establishing collaboration between the various
Gaelic Societies and to hold a conference to create a settled and recognised
system of orthography and grammar. MacBain was less optimistic about getting
consensus noting that 'One humorous gentleman asked, on hearing of the proposed
conference, who would come out of it alive? For we know of friendships having
been broken and professional advancement wrecked over a little Gaelic
apostrophe!' (MacBain 1887b: 45).
While discussions have rarely resulted in formal consensus, there is, on the
evidence of the literature, a general compromise which appears to have the
following features:
Phonographic correspondence is important because it make theorthography 'better' which legitimises the language. This results in 'tidyingup' exceptions to phonographic patterns e.g. tigh > taigh, so > seo.
Uniformity is important. This is because it makes the standard 'better'which legitimises the language. This results in the reduction ofalternatives such as <io> or <ea> in unstressed vowels.
Conservatism, rather than etymology, is important in order to preserveliterary efforts and to not discourage readers.
In Chapters 6, 7 and 8 which look at three particular features — the variation
between <eu> and <ia>, the variation between <s> consonant clusters and the use
of the accents — the decisions about their usages are analysed with this
compromise of ideologies in mind.
93
5 Methodology
This chapter provides an account of the corpus linguistic approach and describes
the electronic corpus used in this study. The broad methodology is outlined here;
more detail on the exact process undertaken for the analysis of the features
investigated in Chapters 6 to 8 can be found in the corresponding chapters. Section
5.1 summarises the approach of corpus linguistics, how it has been previously used
in relation to Scottish Gaelic and why this approach is useful for describing
standardisation in action. Section 5.2 provides the documentation of the corpus
used including a description of its composition in terms of genre, date of
publication and dialect. Section 5.3 describes the use of concordance software to
locate occurrences for quantitative and qualitative analysis.
5.1 The Methodology of Corpus Linguistics
5.1.1 What is Corpus Linguistics?
Advancements in computer technology since the mid-twentieth century have
greatly expanded linguists' use of computer-aided analysis. The field of practice
of computer-aided textual analysis for linguistic research has become known as
corpus linguistics. Corpus linguistics is generally considered a methodology rather
than a discipline of linguistics; indeed, Tony McEnery and Andrew Hardie suggest
that 'the future of the field is in “corpus methods in linguistics” rather than
“corpus linguistics standing independently”' (2012: xiv). The use of corpus data,
and the computer's ability to search for, retrieve, sort and calculate (after Leech
1991), can provide a fast and accurate tool for the linguist. An overview of the
development of the field, which argues strongly for corpus linguistics as 'nothing
but a methodology' can be found in McEnery and Wilson (2001). McEnery and
Wilson characterise corpus linguistics as part of a long-standing debate between
rationalists and empiricists in linguistics as it is based on the observation of
naturally occurring data. McEnery and Hardie (2012) provide a survey of corpus
methods, primarily English language corpus linguistics, and the use of corpora in
studying synchronic and diachronic variation, functionalist linguists and the Neo-
Firthian analysis of meaning in text.
94
While researchers have employed a variety of technologies and approaches, there
are basic underlying methods in common: the corpus researcher aims to process
large quantities of authentic, naturally occurring data in order to obtain objective
evidence and quantitative techniques are applied before qualitative
interpretations are made. It allows linguists to approach language and describe it
better, testing out hypotheses with data on a large scale and discovering new,
unexpected, patterns.
5.1.2 Corpus use in Scottish Gaelic Studies
When it comes to computer-aided research in Scottish Gaelic,
Scottish Gaelic language scholars work in an entirely differentenvironment than English scholars in terms of the resources andworkforce available to them. (Ó Maolalaigh 2013a: 113)
However, the use of machine-readable texts in Scottish Gaelic and Celtic
linguistics is growing as researchers take advantage of new bases of empirical
evidence which were previously hard to access. Existing corpora of, or including,
Scottish Gaelic include the Language Engineering Resources for the Indigenous
Minority Languages of the British Isles (BIML) project headed by Andrew Wilson at
Lancaster University. It contains transcriptions of spoken Scottish Gaelic of
c.17,500 words (comprising a conversation, a university lecture, two sermons and
an informal talk) contributed by Kevin Donnelly of Sabhal Mòr Ostaig. The Gaelic
Text Database GaelDict created by Ciarán Ó Duibhín was first released in 1995
(last release in 1998) which included Irish (mainly Ulster Irish) and Scottish Gaelic
materials with 1,997,043 tokens (running words in text). The Scottish Corpus of
Texts and Speech (SCOTS) at the University of Glasgow has a very small selection
of transcribed conversations (20,687 tokens) in Scottish Gaelic. William Lamb
created a Scottish Gaelic corpus for his 2002 PhD and subsequent monograph in
2008; a benchmark study in register and stylistic marking in modern Scottish
Gaelic. The corpus had a total of 76 texts amounting to 81,677 words and included
a spoken sub-corpus covering Conversation, Radio Interview, Sports Reportage and
Traditional Narrative and a written sub-corpus drawn from the categories of
Fiction, Formal/Academic Prose, News Scripts and Popular Writing (Lamb 2008).
For a useful summary of current resources, see Ó Maolalaigh (2013).
95
Corpus na Gàidhlig and DASG
The most significant corpus project for Scottish Gaelic is the ongoing Digital
Archive of Scottish Gaelic (DASG) project, a British Academy-recognised project
established by Roibeard Ó Maolalaigh in 2006 and founded in order to digitise
valuable parts of the archive generated by the Historical Dictionary of Scottish
Gaelic (HDSG) which was established within the University of Glasgow (1966–97).
Corpas na Gàidhlig was established in 2008 as a constituent project of DASG.
Corpas na Gàidhlig has two aims: to provide a comprehensive electronic corpus of
Scottish Gaelic texts for students and researchers of Scottish Gaelic language,
literature and culture; and to provide the database of texts for Phase II of the
inter-university historical dictionary project, Faclair na Gàidhlig (FnaG -
Dictionary of the Scottish Gaelic Language).18 For a description of DASG and FnaG,
see Pike and Ó Maolalaigh (2013) and Ó Maolalaigh (forthcoming). The digital texts
produced have enabled preliminary linguistic investigation of the corpus, even as
materials continue to be added. A public web interface was launched in 2014 to
make the corpus searchable. This is the first major online corpus of Scottish Gaelic
to be publically available and at the time of writing contains over 10 million
words.19 The author previously used the texts created by Corpas na Gàidhlig for
her MLitt thesis examining possessive forms in modern Scottish Gaelic (Bell 2010).
The texts have also previously been analysed by Ó Maolalaigh (2013) to investigate
singular nouns in Scottish Gaelic and a h-uile (‘every’) with historical implicaitons
(Ó Maolalaigh 2013b). The full possibilities of corpus exploitation to assess
previous linguistic descriptions and to expand our knowledge of Scottish Gaelic
have only just begun to be explored.
5.1.3 Corpus Linguistics for Analysing OrthographicStandardisation
In Chapter 2, the approach advocated by Coulmas was that linguists studying
writing should analyse 'real writing' rather than only considering how writing
systems fit an ideal. The use of an electronic corpus covering the modern period
18 Faclair na Gàidhlig was established in 2003 by an inter-university partnership comprising theuniversities of Aberdeen, Edinburgh, Glasgow, Strathclyde and Sabhal Mòr Ostaig (University ofthe Highlands and Islands), funded by a number of sources: Bòrd na Gàidhlig, the Carnegie Trustfor the Universities of Scotland, the Gaelic Language Promotion Trust, the Leverhulme Trust, theScottish Funding Council and the Scottish Government.19 Available at www.dasg.ac.uk
96
of Scottish Gaelic allows this study to discover how Gaelic spelling has actually
been used in print instead of how it should have or could have been used; to
consider usage as well as principles of orthography. Also, with regards to the study
of standardisation, corpus methods can reveal patterns and norms which help us
document shifting patterns of use, revealing how standards and conventions
evolve and are subsequently either adhered to or rejected by writers. Corpus
analyses can also challenge the 'standard wisdom' within a field. As Raymond
Hickey has observed:
One obvious use of text corpora is to confirm or refute opinions whichhave perhaps been held without any serious degree of questioning byscholars in the field. (Hickey 2007: 120)
As the review of previous studies on Scottish Gaelic orthography in Chapter 3
illustrates, existing discussions have focussed on idiosyncratic forms as useful
evidence sources for historical linguistics, or on contentious features where the
author wishes to advance their own solution to the perceived problem. The broad
patterns contained in the corpus give evidence for wider trends and norms rather
than the idiosyncratic exceptions and outliers as the corpus analyses in the
following chapters show.
5.2 Corpus Documentation
5.2.1 Corpus Construction
The corpus exploited for this research is constructed from the materials created
by the DASG team. Permission and access was given in 2011 for the files which
had, at that time, been digitised and proof-read from the list of texts intended to
form the Faclair na Gàidhlig historical dictionary database. An account of the
digitisation process of the texts can be found in Pike and Ó Maolalaigh (2013).
Designing the corpus
Professors William Gillies and Donald Meek made the initial selection of texts
suitable to form the textual basis of an historical dictionary during Phase 1 of the
Faclair na Gàidhlig project. Dr Catriona Mackie, Research Assistant, researched
and collated metadata of the corpus texts which has also been of great help to
this study. The selection which was made aimed to cover a wide range of texts
97
over three hundred centuries, particularly including influential texts and with a
range of dialect representation, register and topics (see Ó Maolalaigh 2013). It is
therefore a solid foundation for a corpus of written Scottish Gaelic.
Metadata
For each text of the corpus used in the thesis, the metadata of the author (and
editor if transcription was carried out by that person) was collected.
Table 5-1 Metadata features for corpus textsMetadata Feature NotesName of Author In both English and Gaelic formsDate of BirthDate of DeathDialect Both region and parishLiteracy in Gaelic Any relevant information about their educationTitle of Text Short titleDate of Production If known to be different from publicationDate of PublicationBibliographic information Full bibliographic reference
See Appendix 2 for a sample metadata page from DASG of Am Fear-Ciuil (1910).
For some texts, more detailed information than that in the Faclair na Gàidhlig
metadata was gathered. For example the collection of historical essays
Oighreachd is Gabhaltas [Estate and Tenancy] (1980) has several authors and
metadata information details for the author of each essay in the publication was
collected. This meant that each essay could be considered a 'text' with different
dialect information and more complete data for each author could be used to
assess the variety of the corpus contents.
Not all texts were divided by author, however: in Na Baird Thirisdeach [The Tiree
Bards] (1932), the orthography was not determined by the individual poets (all
from Tiree) but by the single editor, Hector Cameron (also from Tiree), and so
there was no value in dividing the texts. The same applied to An t-Oranaiche [The
Songster] (1877) edited by Gilleasbuig Mac na Ceàrdadh [Archibald Sinclair]. The
magazine An Teachdaire Ùr Gàidhealach [The New Gaelic Messenger] (1835-36) is
considered as a whole in this study due to the sheer volume of individual texts and
authors, often using pseudonyms.
98
5.2.2 Corpus Description
This section describes and evaluates the contents of the corpus. The contents of
the corpus can be measured in different ways, one of which is in terms of word
count. The corpus can also be measured by Type Token Ratio (TTR) which indicates
lexical variety. The corpus is drawn from 117 published works between 1750 and
2007.20 There are 251 text files which contain over 4,250,000 words. This section
will detail the composition of the corpus by date, genre and dialect.
Date of texts
The oldest published text in the corpus is Gairm an De Mhoir [The Call of the Great
God] (1750), Alexander MacFarlane's translation of A Call to the Unconverted to
Turn and Live (1658) by Richard Baxter, which was published in Glasgow by Robert
and Andrew Foulis. The most recent published text is Saoghal Bana-mharaiche [A
Fisherwife's World] (2007), an ethnographic study of the life of a fisherwife and
her community through the anecdotes of Bell Ann MacAngus and other Gaelic
speakers from the fishing villages of Easter Ross by Seòsamh Watson, Professor
(Emeritus) of Modern Irish Language and Literature at University College, Dublin.
In Table 5.2 below, the number of tokens per decade is shown. As this study is
interested in orthography, texts have been counted here under their orthographic
date: e.g. Tuath is Tighearna [Tenants and Landlords] (1995) is a collection of
nineteenth-century poems; however, its orthography was modernised for its
edition of 1995. Thus, it is included in the 1990s word count. Similarly, the editor
of Gaelic Songs of Mary MacLeod (1934) modernised the spelling of the language
originally dating c.1616-c.1707 to the 'correct modern standards' (Carmichael
Watson 1934: vii) although there are variant readings in the footnotes. This is the
same for An Clarsair Dall [The Blind Harper] (1970), Òrain Iain Luim [The Songs of
Iain Lom] (1964) and Sìlis na Ceapaich [Silis of Keppoch] (1972) where the
orthography is generally that of the respective twentieth-century publication date
despite some older forms appearing on occasion. In the case of Eachann Bacach
and other MacLean poets (1979), two poems, 'Rainn' by Maighstir Seathan and
'Aindra Mc Ghiléoin cecinit' by Anndra Mac an Easbuig have been presented in the
original spelling from their publication in 1707 (in Lhuyd's Archaeologia Britannica)
20 See Appendix 1 for a detailed table of metadata information on texts.
99
while 'All the other poems I have attempted to present in an acceptable modern
spelling' (Ó Baoill 1979: viii).
Table 5-2 Corpus data by decade
Decade Tokens Types TTR%Corpus
% byCentury
1700s 4,468 1,871 41.87 0.121750s 92,295 13,357 14.47 2.161760s 240,774 11,217 4.85 5.631770s 15,655 2,659 17.23 0.361780s 0 0 01790s 9,976 2,050 20.72 0.23 8.391800s 108,938 11,014 10.11 2.541810s 69,372 10,206 14.82 1.621820s 87,648 10,433 12.00 2.051830s 205,342 22,641 11.02 4.801840s 164,626 22,631 13.80 3.851850s 0 0 01860s 47,887 9,086 18.97 1.121870s 141,539 20,991 14.83 3.311880s 51,041 13,000 25.46 1.191890s 341,594 73,094 21.39 7.99 28.501900s 58,722 12,816 21.82 1.371910s 251,756 47,745 18.96 5.891920s 90,957 8,865 9.74 2.121930s 362,785 71,664 19.75 8.491940s 43,150 4,896 11.34 1.001950s 18,180 6,523 35.88 0.421960s 168,388 27,720 16.46 3.941970s 813,299 125,226 15.39 19.031980s 328,104 58,236 17.74 7.671990s 327,369 68,791 21.01 7.66 57.632000s 233,317 27,069 11.60 5.46 5.46TOTAL 4,272,714 681,930
Nearly all decades are covered by the corpus texts with the exceptions of the
1780s and 1850s. There is a substantial difference in the level of representation
by decade which ranges from 0.12% to 19.03% of the total corpus by token. The
majority of the corpus, just over 57%, is from the twentieth century. Nearly a fifth
of the whole corpus, 19%, dates from the 1970s.
Genre
In Table 5.3 below, the corpus is divided into prose or verse and into four genres;
Religious: incl. sermons, catechisms Imaginative: prose fiction, modern poetry
100
Expository: description, instruction, analysis, biography Traditional: folktales, songs
Certain texts contain a number of prose and verse elements that were not easily
divided by genre and thus 6% of the corpus is counted as mixed prose/verse here.
Table 5-3 Corpus data by genreProse Verse Prose/Verse % of corpus
Religious 479,021 176,322 0 15.3%
Imaginative 754,375 73,111 48,577 20.5%Expository 876,000 0 1,659 20.5%Traditional 515,895 1,129,460 95,366 40.7%Mixed 0 0 124,760 3%Total tokens
% of corpus
2,625,291
61.4%
1,378,893
32.3%
270,362
6.3%
100%
The high percentage of verse in the corpus reflects the dominance of poetry in
the Gaelic literary tradition and in Gaelic publications.
Dialect
Scottish Gaelic dialects are considered to be particularly rich in variety, and at
the same time, similar enough for the speaker population to consider it to be one
language. Table 5.4 below shows the distribution of dialects in the corpus. This
unfortunately cannot be compared to real-world data without a large scale
analysis of census data which is not available. The regions and districts used here
are based on the Survey of the Gaelic Dialects of Scotland (SGDS). The rationale
behind the choice of regions is explained as follows:
For the islands, areas are distinguished on a geographical basis, butsocio-geographical and linguistic criteria are used in establishing themainland districts. (Ó Dochartaigh 1997: 69)
The first division is across Scotland, using the 'Highland Line' to delimit the Gaelic-
speaking area. The mainland areas are then divided by county and then by
'traditional' districts. While SGDS has just one region of 'Mainland Inverness-shire',
I have divided this into North & East Inverness-shire and West Inverness-shire for
better specificity and so that the greater percentage of island dialects of Skye &
Raasay are more clear. Conversely, I have grouped some of the smaller
101
neighbouring mainland areas together.21 The spread of dialectal representation in
the corpus is as follows:
Table 5-4 Corpus figures by dialectRegion & District Tokens % of Corpus Regional
% of corpusWestern Isles 38%St Kilda 64,326 1.5%Lewis 741,795 17.3%Harris 314,111 7.4%North Uist 134,032 3.1%Benbecula 2,273 0.1%South Uist 288,786 6.7%Barra 84,288 1.9%Sutherland & Caithness 21,265 0.5 0.5%Argyll 10.7%Coll 21,503 0.5%Colonsay 2,615 0.1%Islay 35,652 0.8%Jura 51,688 1.2%Mid-Argyll 3,076 0.1%Mull & Iona 93,492 2.2%Tiree 107,354 2.5%Appin & Benderloch 139,318 3.3%Ross-shire 2%Easter Ross 72,767 1.7%Wester Ross 10,934 0.3%North & East Inverness-shire 0.7%North Inverness-shire 0Badenoch 7,206 0.2%Strathspey 21,856 0.5%West Inverness-shire 12.9%Skye & Raasay 394,109 9.2%Small Isles 68,876 1.6%Lochaber 44,164 1.1%Morvern 16,243 0.4%Sunart & Ardnamurchan 4,556 0.1%Moidart 24,280 0.5%Perthshire 198,467 4.7% 4.7%Unknown 229,913 - 5.4%Various 1,002,237 - 23.4%Learner* 71,532 - 1.7%
Corpus Total 4,272,714 - 100%
* Learners with no known specific dialect affiliation.
21 They are Glenelg & Knoydart, Morar & Arisaig and Sunart & Ardnamurchan.
102
Almost a third of the corpus does not have a dialect attributed: either because
the information about the author is lacking, or because the text is in a magazine
or collection format where many authors are represented in the text.
The majority of the texts represent the Gaelic of the Isles: 56.1% of the corpus
when counting the Western Isles, Skye & Raasay and the Inner Hebrides.
The type/token ratio
There are 4,274,546 tokens (individual running words in text) and 678,925 types
(unique word forms) in the corpus.
The Type/Token Ratio (TTR) is calculated as the number of types divided by the
number of tokens, and is an indicator of how lexically varied the vocabulary in the
corpus is. The result is often multiplied by 100 to express the type/token ratio as
a percentage (as it will be in this thesis); the closer the ratio is to 1:1, (or 100% if
expressed in percentages), the more varied the vocabulary is. This allows for
vocabulary variation to be measured between corpora and texts. The two
published works with the highest TTR are both wordlists produced by Museum nan
Eilean; Croitearachd [Crofting] (1991), 52.42% and Iasgach [Fishing] (1991),
52.79%. Due to their nature as wordlists, repetition is minimised and the texts
themselves are short: 394 tokens and 249 tokens respectively. The published text
with the lowest TTR is Tiomnadh Nuadh [New Testament] (1767) at 4.66%. This is
likely due to it being the longest text in the corpus, with 240,774 running words
(tokens), where repetition becomes more likely.
Representativeness
A corpus is only a sample of the language that is under analysis. If any
generalisations from that sample are to be made, it is crucial to understand how
representative that corpus is of the language overall. Biber et al. (1998) and
McEnery and Wilson (2001) amongst others have noted the importance of defining
the limits of the corpus under study; without this clarification results can be
misinterpreted as being representative of a much broader language sample than
is actually evidenced in the data. A key problem for Gaelic corpus linguistics is
that we are still relatively unsure of the total composition of the written language.
With the worked carried out by DASG, however, we can be confident that the
103
current corpus resources for Scottish Gaelic have an expansive breadth of
authorship, dates, genre and dialects.
Geoffrey Leech (1992) argued that corpora provide a powerful methodology from
the point of view of the scientific method as they are open to objective
verification of results. Or, as noted by McEnery & Wilson,
Corpus-based observations are intrinsically more verifiable thanintrospectively based judgements. (2001: 14)
With this in mind, the principal strength of the corpus used here is its breadth
over a wide time-span and the substantial amount of objective data it provides
from published texts printed between 1750 and 2007. It provides a contrast to the
more introspective-based evaluations of Scottish Gaelic orthographic history. The
corpus used here prefers complete texts over samples of texts as this results in a
data set of orthographic usage which is as wide as possible. The published nature
of the works also means that they are, in the main, written and/or edited by
literate, well-educated writers. This lends itself to the primary research goals
which are to investigate standardisation at work in orthography.
As with almost all other corpora, the corpus used here is partial in the sense that
it is an incomplete record of the language (as natural languages are not finite)
(see McEnery & Wilson 2001: 10). The corpus does not contain all written materials
in Gaelic, thus, the earliest or latest example of a spelling in the corpus does not
attest necessarily to the earliest or latest form in written Scottish Gaelic. The
corpus does not provide data on speech, unpublished texts, or informal writing
(personal letters, emails), as the corpus comprises published materials only. The
corpus is also limited in its coverage of children's literature, Gaelic periodicals,
journalism and news reports.
5.3 Analysis
5.3.1 WordSmith Tools
The analysis of corpus data was first carried out using the WordSmith Tools
concordancer, Version 5 (Scott 2008). The concordancer allows for instances to be
extracted as .xls files where occurrences can then be tagged and counted. Despite
104
major sections of the texts in other languages (mainly English) being removed,
there are still quotes, notes and references in other languages, particularly English
and Irish. Also, older and arcane spellings in Scottish Gaelic are sometimes found.
By seeing occurrences in context in the concordance data, irrelevant occurrences
could be removed before analysis began, i.e. sections that were not in Gaelic (e.g.
English, Irish, Latin); printing errors (e.g. an-schocair, nsch eil, for a-shocair and
nach eil respectively) and loan words with no orthographic change (e.g. pistol,
elastic, asbestos, scrap). Occurrences within their immediate context also meant
homographs were more easily semantically separated and data could be
lemmatised in order that variations from mutation, plurals, slenderisation etc.
could be counted together.
Figure 5-1 WordSmith concordance
Searches can be carried out on WordSmith using exact matches or by using * as a
wildcard for any letter. Thus, a search for *ia* returns all words starting with,
containing or ending with <ia>. WordSmith can then sort the results
105
alphabetically. Once WordSmith had been used thus to retrieve the relevant
occurrences, the data was exported as an Excel file (.xls).
Figure 5-2 Sample of spreadsheet data
Figure 5.2 shows an extract of an annotated spreadsheet. The filenames were
created with the format of date preceding the title. This meant that results could
easily be seen in chronological order by using Excel's sort function on the 'Text'
column containing the filenames. In the column 'Set', the occurrence is marked
with the feature to be analysed, in this case initial, medial or final <sc>. In the
'Tag' column, each occurrence is assigned to a lemma regardless of case, mutation,
pluralisation etc. Any spelling variation other than the variation the one being
analysed is ignored e.g. maicroscop and miocroscopach in this example. Further
columns can include information on sub-divisions of the texts, page references to
the publications, and any other notes.
In another tab on the spreadsheet, a formula (using COUNTIFS) was entered to
calculate totals of texts and lemmas in each text as shown in the following figure.
106
Figure 5-3 COUNTIFS formula in Excel
The data collated in this way could then be used for charts and graphs. Patterns
and exceptional occurrence could then be investigated more closely by examining
texts individually and qualitatively.
5.4 Summary
This chapter has outlined the choice of corpus linguistics as a tool for investigating
the standardisation of orthography. The digitisation projects of Faclair na Gàidhlig
and DASG which provided the texts for this study's corpus has been described. The
contents of the corpus provide a wide-ranging overview of written texts in Scottish
Gaelic in the modern period which form the evidence basis for the description of
changing spelling conventions.
107
6 Acceptable Variation: <eu> and <ia>
In this chapter I trace how the preserved long /eː/ and broken long /ia/, have
been transliterated in the orthography as <eu> and <ia>. As this feature indicates
dialectal difference, it is a useful one with which to consider how Scottish Gaelic
orthography has reconciled the competing ideas of the phonographic ideal (leading
to the representation of different varieties) and the push of standardisation
towards uniformity without variation.
Section 6.1 summarises the breaking of the long /eː/ and its significance to
Scottish Gaelic dialect studies. Section 6.2 summaries how previous prescriptive
and descriptive texts have aimed to cover the phenomenon and which arguments
have been used for and against variation. Section 6.3 then analyses the corpus
data to see how usage has matched or diverged from description. It focuses on
individual word pairs to see when each spelling has been most in use over the
period contained in the corpus. Although the corpus does not cover texts before
1750, some early uses of <ia> in MSS have been identified where they are
reproduced or quoted in later texts.
As the primary interest is in the standardisation of the use of these two variants,
other forms, such as <éa>, which have been used occasionally over the same
period will not be considered here.22 It will also be limited to variation where the
<eu> form is part of a monosyllable, e.g. beul. Compound words have different
factors to consider, such as whether the vowel is diphthongised when the stress is
moved; compare beul, beul-aithris.
6.1 The Breaking of Long /eː/
The occurrence of /eː/ or /ia/ in words with historical /eː/ is a distinguishing
feature separating Scottish Gaelic dialects. The feature is derived from the
'breaking', or diphthongisation, of Old Gaelic stressed long é, [eː], before non-
palatal consonants. According to the Rev. Charles M. Robertson (1885-1927), who
authored some important early papers on Scottish Gaelic dialects, the breaking
can be found in 110 words in Scottish Gaelic (1907: 89). In the Survey of the Gaelic
22 For example in Carswell (1567) the norm is <é> with occasional uses of <éu> and <éa>including béul (Thomson 1970: xiv).
108
Dialects of Scotland (SGDS), the questionnaire included, a' cheud 'first', deug
'teen', beul 'mouth', feur 'grass', sgreuch 'screech', gun deunadh 'that he would do',
Seumas 'James', sè 'six' (Jackson 1968: 66). It was the only feature that the Survey
leader, Kenneth H. Jackson, published a description of using the SGDS data (Ó
Dochartaigh 1997: 53). From the data, Jackson (1968) identified a zone where
words with long /eː/ are best preserved which stretches from the south west to
the north east of Scotland (Argyllshire across to Perthshire, Strath Spey and north
to Easter Ross and Sutherland). In older texts, the areas are referred to as
'northern' and 'southern' although modern dialectology describes these as central
and peripheral areas (following Jackson 1968: 67). The central areas have tended
towards the diphthong pronunciation while peripheral areas have preserved the
older pronunciation e.g. beul, 'mouth'; Uist /bial/ vs. Arran /bɛːl/.
6.2 <eu> and <ia> in the Literature
6.2.1 The Codification of <eu>
This section shows how the southern (or peripheral) dialect writers who codified
the norms in the early eighteenth century were aware of the dialectal variant <ia>
but did not consider it an acceptable spelling. The norms which were created at
the end of the eighteenth and start of the nineteenth centuries were set by writers
from peripheral areas, particularly Perthshire and Argyllshire: for example, the
Rev. Dr John Smith (1747–1807), a native of Glenorchy; Rev. James Stewart of
Killin (1700-1789), originally from Glenfinlas in the Trossachs; and Robert
Archibald Armstrong (1788–1867), born at Kenmore, Perthshire. In the early
nineteenth century, we can find references to the <eu>/<ia> contrast in guides to
pronunciation. The Rev. Dr Alexander Stewart acknowledges it in his grammar:23
The sound of eu is like (2) e alone: long, as, 'teum' to bite, 'gleus' trim,entertainment. One of the most marked variations of Dialect occurs inthe pronunciation of the diphthong eu; which, instead of beingpronounced like long e, is over all the North Highlands commonlypronounced like ia; as, 'nial, ian, fiar', for 'neul, eun, feur.' (Stewart1812: 7-8)
23 It is not noted in the first edition of 1801 but appears in the second edition in 1812.
109
The Rev. Dr Alexander Stewart (1764-1821) was a native of Blair Atholl, Perthshire
and later minister of Dingwall, Easter Ross (1805-20) which made him a speaker
of a peripheral dialect but familiar with a northern, central one (Meek 1993c:
792). A contemporary of Stewart's, Alexander MacLaurin (1740-1820) was from
Comrie, Perthshire and lived in Edinburgh where he would have met Gaelic
speakers from varied backgrounds (Thomson 1994: 180). He also noted the
dialectal variation in his school guide for the SSPCK:
In the North Highlands, eu is pronounced like ia; as nial for neul, fiarfor feur, &c. (McLaurin 1816: 54)
In the mid-nineteenth century, Archibald Farquharson named the <eu> convention
explicitly as the 'standard':
Besides these, there are many words where eu may be changed into ia,as feur, geur, neul (grass, sharp, cloud), fiar, giar, nial. The former isthe standard Gaelic, but the latter is more common in the west andnorth. (Farquharson 1868: 23)
Farquharson (1801-1878) was another Perthshire native whose job as a minister
took him to different districts including many years in Tiree (MacLean 1915: 116).
Although <ia> is not presented in the above examples as a valid orthographic
alternative, these writers used the digraph <ia> as a guide to pronunciation which
they clearly expected to be understood. This is because <ia> was already in use in
some other words to represent the historical diphthong /ia/ in words such as
biadh, 'food', grian, 'sun'. These are usually realised as /iə/ and thus generally
contrasts with broken é which is usually realised as /ia/. By the end of the century,
it was recognised by MacBain that the codification of /ɛː/ as <eu> was as a result
of the peripheral dialects' influence on the 'literary language':
The crucial distinction [between north and south] consists in thedifferent way in which the Dialects deal with é derived fromcompensatory lengthening; in the South it is eu, in the North ia (e.g.,feur against fiar, breug against briag, &c.) […] The Southern Dialect ispractically the literary language. (MacBain 1911: vii)
Later in the nineteenth century and into the twentieth, this standard spelling was
challenged by writers of northern dialects as the following section describes.
110
6.2.2 Dialects and Acceptable Variation
In Section 2.2.1.2, it was noted that there are models of orthography where
variation can exist within a standard. In Section 4.3, the various positions that
Scottish Gaelic writers have taken on the suitability of dialectal variation was
documented, particularly the argument that dialect variant spellings were
appropriate in poetry and folklore which I have termed MacKinnon's paradigm. The
<eu>/<ia> distinction is so iconic that it was commonly referenced when discussing
the nature of acceptable variation.
In MacKinnon's article 'On the Orthography of Scottish Gaelic', published in The
Celtic Review in 1909, we can see the elucidation of the ideology of the standard
in relation to dialects. MacKinnon was intolerant of 'localisms' and he derided 'the
tendency to the excessive use of localism in the orthography of Scottish Gaelic
which threatens to bring our limited literature into disrepute' (1909: 13).
However, he considered that a less 'excessive' use in certain areas such as folklore
was acceptable:
When Mr. Campbell [John F. Campbell] began to collect his PopularTales it was a sound literary instinct that suggested the presentment ofthem in dialectal form. (1909: 13)
When he goes on to specify two acceptable exceptions to the standard, it is the
variation of <eu>/<ia> that he uses to illustrate his point.
(1) In writing lyric verse the ring of the line must be preserved. Thepoet will use, and ought to use, not only such double forms as eun andian, beul and bial, but all local sounds that lend grace and melody tohis rhymes.
(2) When illustrating dialect, or registering dialectal material forlinguistic and historical purposes […] (1909: 17)
This paradigm that MacKinnon sets out had been alluded to by earlier writers.
Many examples can be found providing evidence that this was a deliberate
practice, not a mark of partial literacy or a misunderstanding of the standard
spelling conventions. In each of the following quotes, the distinction between
<eu> and <ia> is the example given by the writer when justifying their spelling
approach. In his Collection of Gaelic Proverbs and Familiar Phrases (1881),
111
Alexander Nicolson also argued for the appropriateness of variant spellings in
terms of MacKinnon's paradigm:
(...) an occasional divergence from the canonical norm, and even variedspellings of the same word, have seemed to me not only excusable butdesirable. The phrases in which these words occur belong to thesimplest vernacular forms of speech, and ought to be so given as torepresent faithfully the varieties of phrase and pronunciation foundamong Gaelic-speaking people. The greater part of the two thousandthree hundred sayings here first collected were received in MS., mostlyfrom good Gaelic scholars, who spelled sometimes in different ways.Among these varieties of spelling are béul and bial, bréug and briag,féur and fiar, sgéul and sgial, rìs and rithist, &c. To adhere uniformlyto any of these would sometimes spoil the rhyme or rhythm on whichthe charm of a proverb often depends. (Nicolson 1881: xi)
The issue of rhyme and rhythm is also the important factor for the author of the
next example which comes from the introduction to a set of folk tales in the Celtic
Magazine. The author's interest here is in stories and songs 'in the pure dialect of
the district' (Alastair Og 1877: 28). Pre-empting any criticism for his use of non-
standard forms, his argument for the use of <ia> is not to challenge the standard
but to sustain rhyme and harmony:
Certain very expressive words peculiar to the district will be noticed,and it will be remarked that the words beul, meur, feuch, and suchlike, are pronounced bial, miar, fiach, and so on. Such words as thesemay easily be altered in prose writings, without any injury to the text,but it is impossible to do so in poetry, the sound being so very different,without altering the harmony and consonance of the piece. This willaccount for our giving the Gaelic Songs throughout the Ceilidh in thedialect of the district in which they were composed, and our answer toany who may consider the orthography faulty and not in accordancewith the now almost universally received standards. (Alastair Og 1877:28)
The variation is also considered an advantage to poets. By having two forms
representing two sounds, <eu> for /eː/ and <ia> for /ia/, the poet has more
flexibility in rhyming choices (rhyme in Gaelic poetry is created mainly by
assonance). In his notes on the Harris dialect of Iain Morison (Gobha na Hearadh),
George Henderson also describes how /ia/ accords with Morison's dialect, yet in
his poems he can 'take advantage' of both.
1. The dialect of the poems is distinctly the northern; it has ia asequivalent for the ē (long open e), which arose through compensatory
112
lengthening, e.g. ciad, nial. Both forms, northern and southern, aretaken advantage of, for the sake of rhyme. (Morison 1896: 348)
Another example of this function in poetry, where both forms are available to the
poet, dates back to the eighteenth century. Angus MacLeod in the edition of The
Songs of Duncan Bàn MacIntyre (1952) observes that MacIntyre alternates between
both vowels 'as rhyme demands'.
and the poet clearly used both -ia- and -eu- forms as rhyme demands,e.g. both ciataichead and ceutach. (MacLeod 1952: xliv)
MacIntyre himself was illiterate and the first edition of his work was produced in
1786 by Rev. Dr John Stuart of Luss (1743-1821) who subsequently took part in the
translation of the Old Testament (1801) and the revision of the New Testament
(1796).
Resistance to the <ia> variation can also be found, however. In a review of
Sgialachdan á Albainn Nuaidh [Stories from Nova Scotia] (1970) by C.I.N. MacLeod,
Seumas Robasdan criticises MacLeod's use of <ia> saying:
Gheibhear -ia- air a litreachadh an àite -eu- gu cunbhalach anns anleabhar so, mar shamhladh: bial, ian, nial. Tha fios gu bheil an dòigh-litreachaidh so a' co-fhreagairt ris mar tha na faclan air a labhairt, achcuiridh mi an aghaidh an fhasain so airson na reusain a leanas: (1) 'sabhaist dhuinn -ia- a litreachadh ann am faclan leithid liath, ciar, grian,air a bheil fuaimneachadh /iə/ (2) a thaobh fuaim is ciall le chéile,cuiridh mi deafar eadar: feur ("feur gorm"), fiar (cam); feuch (dh'fheuchmi am biadh"), fiach ("fiach trì tasdain"); seun (cuimhneachan air a'bheannachadh a ghléidheas duine an àm cunnart), sian ("na seachdsiantan"), (3) gheibhear facal leithid 'geur' air a bheil fuaimneachadh/eː/, ann an cuid cheàrnaidhean air Ghaidhealtachd co-dhiù; chafhreagradh an litreachadh 'giar' idir ris mar a tha am facal air a labhairtanns na ceàrnaidhean sin." (Robasdan 1970: 383)
–ia- is found spelt in place of –eu- consistently in this book, forexample: bial, ian, nial. Certainly this spelling corresponds to how thewords are said, but I am against this fashion for the following reasons:(1) we are used to using –ia- in spelling words such as liath, ciar, grian,which are pronounced /iə/ (2) in terms of sound and meaning together,I distinguish: feur (grass), fiar (crooked); feuch (try), fiach (worth);seun (a charm), sian (violent weather), (3) a word like 'geur' ispronounced /eː/ in at least some parts of the Gaidhealtachd; thespelling 'giar' would not correspond to how the word is said in thoseareas.
113
Despite the context of folkloric stories in the text, Robasdan does not accept this
variation but makes a case for preserving the <eu> spelling. Also in the 1970s,
Donald Archie MacDonald made it clear in his introduction to Pàdruig Moireasdan's
Ugam agus Bhuam (1977) that he wished to represent Pàdruig's speech.24 However,
there are nine specified examples where he chooses to use a more conventional
spelling over one which is more dialect specific, and this list includes using <eu>
in place of <ia>.
7) Far an can sinn, am bitheantas, bial, sgial, etc. tha mi air a bhith'sgrìobhadh beul, sgeul, etc. (1977: 19)
Where we say, generally /bial/, /sgial/, etc. I have written beul, sgeul,etc.
The nature of variation between <eu> and <ia> as a dialect marker makes the
feature a useful one to illustrate the relationship between standard and non-
standard forms. As section 6.3 will show, the corpus examples of <ia> primarily
fall into the two contexts in MacKinnon's paradigm: for rhyme in poetry and for
texts with a particular linguistic or dialectal interest.
6.2.3 <eu> and <ia> in GOC
The GOC report of 1981 recognises <ia> as an existing variant form of <eu>:
In the case of words spelt with eu (for example, beul, ceud, meud)there would be no great gain in making a change in the spellinggenerally (to bial, ciad, miad, for example, though such spellings arecommonly enough found in contemporary sources and have to berecognised, therefore, as occurring forms). (SEB 1981)
Its wordlist, however, lists only 'beul', 'beul-aithris' and 'beulaibh' and not their
<ia> forms. While it does not, then, explicitly prohibit the use of <ia>, neither
does the report support or encourage its use. What GOC 1981 does recommend is
the use of the two spellings to distinguish between homophones:
3.3 When there are homophones, however, these should bedifferentiated in spelling where such differentiation is feasible. For
24 'Am bitheantas, cho fada agus a tha e ann an comas a’ litreachaidh sin a dheanamh, tha mi airmo dhìchioll a dheanamh air cainnt Phàdruig fhéin a thoirt a follais air an duilleig mar a tha i airan teip.' (1977: 18) (Generally, as far as it is possible for the spelling to do so, I have tried tomake Padruig's own speech as clear on the page as it is on tape.)
114
example, in the case of ceud = 100 / ceud = 1st, ceud = "100" shouldbe retained and "1st" should be written ciad; and mìos = "month" whilethe word for "basin" should be written mias; feur = "grass"; fìor ="true"; fiar = "squint". (SEB 1981)
By the 2005 recommendations however, 'eu rather than ia' was explicitly the
recommended form (SQA 2005: 4A) and this was repeated in GOC 2009. The use
of the two forms to minimise ambiguity by distinguishing between homophones
was maintained in the later reports (see Section 6.3.5 feur vs fiar).
6.3 <eu> and <ia> in the Corpus
6.3.1 Lemma Frequency
In Table 6.1 below, the frequency of occurrences for the most common pairs of
<eu>/<ia> variation across the corpus are shown. The data in this table does not
distinguish semantic differences such as ciad 'first / ceud 'hundred', or feur 'grass'/
fiar 'crooked' (6.3.5 shows differentiation between feur/fiar).
Table 6-1 Frequency of eu/ia word pairs in the corpus<eu> Freq. <ia> Freq.
ceud 1,851 ciad 1,008beul 1,481 bial 252sgeul 1,224 sgial 104geur 881 giar 21deug 619 diag 163meud 440 miad 52feur 350 fiar 147breug 206 briag 45deur 184 diar 43meur 154 miar 49neul 131 nial 42freumh 87 friamh 27
The general picture for <eu>/<ia> variation in the corpus is that <eu> is more
common in all lemmas. In the following sections, the word pairs of beul/bial,
sgeul/sgial, geur/giar, and feur/fiar will be discussed in turn.
115
6.3.2 ceud vs ciad
The lemmas ceud and ciad can both refer to either 'first' or 'hundred'.25 The GOC
1981 report codified a distinction between them, assigning ceud as 100 and ciad
as first. There are 714 occurrences of ceud (as 100) and 202 occurrences of ciad
(as 100) in the corpus. There are 953 occurrences of ceud (as first) and 773
occurrences of ciad (as first).
In the eighteenth century texts, ceud is the norm for both meanings as the
following chart shows.
Figure 6-1 Frequency of ceud/ciad in 18th-century corpus texts
In Gairm an De Mhoir (1750), neither form is used, cead is used for both meanings.
MacMhaighstir Alasdair also mainly uses cead for 'first'. 26 On one occasion,
however, he uses ciad (first) in the poem 'Moladh an ughdair do 'n tsean chánoin
Ghailic':
’S a rèir Mhic Comb,An t-ùghdar mòr ri luaidh,’S i ’s freamhach óir,’S ciad ghrámair glóir gach sluaigh (Mac-Donuill 1751: 7)
As the next section will show, Ais-éiridh (1751) is also the first corpus text to use
bial. This confirms the expectations of where <ia> would be found according to
dialect studies and MacKinnon's paradigm: MacMhaighstir Alasdair's dialect of
Moidart is within the central dialect area and he is using <ia> in verse, although it
25 Dwelly redirects ciad to the headword ceud but also offers a second definition: (CR) 'Opinion,impression – Strathtay. Ghabh mi droch chiad dheth, I formed a bad opinion of him' (1911 s.v.ciad) sourced from the Rev. C. M. Robertson. This use has not been found in these corpusresults and appears to be obsolete.
26 e.g. 'Chead Choirnel a bha ' g a Ph — — sa' (Mac-Donuill 1751: 112)
0 20 40 60 80 100 120 140
Gairm an De Mhoir (1750)Ais-éiridh (1751)
Tiomnadh Nuadh (1767)Leabhar Ceasnuighe Aithleasuighte…
Saighdear Criosduidh (1797)
ceud 1st
ceud 100
ciad 1st
ciad 100
116
does not appear to be for rhyme in this instance. The remaining eighteenth-
century texts use ceud for both 'first' and 'hundred'.
In the nineteenth century there are many more occurrences of ciad but there is
no distinction in meaning between the two forms and their two meanings.
Figure 6-2 Frequency of ceud/ciad in 19th-century corpus texts
The first nineteenth century text to use ciad in significant numbers is Patrick
Turner's poetry collection Comhchruinneacha do dhain taghta Ghaidhealach
(1813). Turner was a native of Cowal who travelled throughout the Highlands
collecting poetry (MacLean 1915: 361). His use fits with MacKinnon's paradigm as
it is poetry and ciad is often (but not always) used for rhyme as in the following
example:
Gu ’n robh mais ann a t-fhiamh,’Sin a’s tlachd ort measg chiad,
117
Rud nach cuala’ mi riamh air triùir. (1813: 129)
This is, however, the only <ia> form that Turner uses. As the next sections will
show, ciad is substantially more common in the first half of the nineteenth century
than the other lemmas with <ia>. This is reflected in Armstrong's Dictionary (1825)
which does not have headword entries for <ia> spellings generally but does have
one for ciad. Armstong gives its meaning as 'A hundred. More frequently written
ceud' (Armstrong 1825 s.v. ciad). Ciad is the first of the broken long e <ia> lemmas
to be recognised.
Most of the times where ciad appears in the nineteenth century it is in books of
poetry, although even in these contexts, ceud is the norm; i.e. in Orain Ghaelach
(1801), Sar-Obair nam Bard Gaelach (1841), Laoidhean agus Dàin (1868). All of the
examples of ciad, whether as 'first' or 'hundred', all appear in verse until An
Teachdaire Ùr Gàidhealach (1835-6). Here, is it used in some contexts that fit
within MacKinnon's paradigm as 'dialect materials' such as prose dialogues e.g.
'Duilleag nam Balachan' by Taracul Udlaidh (1835: 172) but occasionally in other
prose contexts such as news and letters; e.g. 'Guth O MhacTalla' (1835: 235)
The <ia> form continued to be used with no apparent distinction between
meanings for the rest of the nineteenth century along with the <eu> form. In the
last decade of the century, the <ia> form is the most common form of ciad in the
corpus. In the first couple of decades of the twentieth century, however, ciad,
becomes less common despite its use in Lòchran an Anma (1906) and An t-Ogha
Mor (1913), both of which as will be seen in following sections, only use <ia>
forms. This reduction in prominence of ciad may have been influenced by its
codification as ceud in MacBain (1896 & 1911) and Dwelly (1911).
118
Figure 6-3 Frequency of ceud/ciad between 1900 and 1969
Neither Dùn-Aluinn (1912) nor Aig Tigh na Beinne (1911) uses any <ia> of the
lemmas under analysis. In contrast, Lochran an Anma (1906), An t-Ogha Mor (1913)
and Litrichean Alasdair Mhoir (1932) use ciad exclusively: these three texts are
frequent users of <ia> forms as the following sections will illustrate. By the 1960s,
ciad is beginning to establish itself as the norm over ceud. This change can be
followed in the writings of Iain Mac a' Ghobhainn (Iain Crichton Smith) who has
several texts in the corpus. In his collection of stories Bùrn is Aran (1960), he only
uses ceud, but in An Dubh is An Gorm (1963), ciad is more common. By the
publication of his novella An t-Aonaran in 1976, he uses ciad exclusively.
119
Figure 6-4 Frequency of ceud/ciad between 1970 and 1980
The decade preceding the first GOC reforms shows a mixed picture but with ciad
appearing in even more contexts outwith MacKinnon's paradigm such as the essays
in Aitealan Dlù is Cian (1972) agus Criomagan Ioma-dhathe (1973). Most
significantly, however, is the first text to make a clear distincion between ciad
(first) and ceud (hundred): Bith-eòlas (1976). This biology textbook was translated
by Ruaraidh MacThòmais [Derick Thomson] who later chaired the GOC sub-
committee. The adoption of this distinction is clearly seen in the texts published
from 1982 (see folllowing chart).
0 20 40 60 80 100 120 140
Briseadh na Cloiche (1970)Luach na Saorsa (1970)An Clarsair Dall (1970)
Tir an Aigh (1971)Lus-Chrun a Griomasaidh (1971)
Saoghal an Treobhaiche (1972)Aitealan Dlù is Cian (1972)
A' Bhratach Dhealrach (1972)Silis na Ceapaich (1972)
Suathadh ri Iomadh Rubha (1973)Nach Neònach Sin (1973)
Criomagan Ioma-dhathte (1973)An Aghaidh Choimheach (1973)Creach Mhor nam Fiadh (1973)
Fo Sgail a' Swastika (1974)An t-Aonaran (1976)
Bith-eòlas (1976)Gàidhlig ann an Albainn (1976)
Ugam agus Bhuam (1977)Hebridean Folksongs Vol. 2 (1977)
Orain Dhonnchaidh Bhain (1978)Eachann Bacach (1979)
Deireadh an Fhoghair (1979)Oighreachd is Gabhaltas (1980)
ceud 1st
ceud 100
ciad 1st
ciad 100
120
Figure 6-5 Frequency of ceud/ciad post-GOC
The distiction is adhered to decisively. The only exceptional occurrences are in
Sgrìobhaidhean Choinnich MhicLeòid (1988) and Hiort (1995). The writings of
Kenneth MacLeod were edited by T. M. Murchison in Sgrìobhaidhean Choinnich
MhicLeòid (1988). While the spelling is generally modernised, the small number (4
occurrences) of ciad, 'hundred', are likely to be old spellings from MacLeod's
original texts. There are 3 occurrences of ciad, 'hundred', in Hiort (1995) which
are not from older quotations or in poetry, but appear to be simple errors. Overall,
the codification of the semantic difference into two existing variant spellings has
been a successful reform.
6.3.3 beul vs bial
The difference between beul and bial is only one of dialectal pronunciation. Both
have the semantic meaning of 'mouth' or 'opening'.27 There are 1,481 occurrences
of beul and 252 occurrences of bial in the corpus. In the eighteenth century texts,
it is beul which occurs predominantly.
27 Dwelly has an entry for bial meaning 'water', however this is tagged as obsolete.
0 10 20 30 40 50 60 70 80 90
Hebridean Folksongs Vol. 3 (1981)Air Mo Chuairt (1982)
Creachadh na Clàrsaich (1982)Suileabhan (1983)
Seann Taighean Tirisdeach (1986)Air Druim an Eich Sgiathaich (1987)
Sgrìobhaidhean Choinnich MhicLeòid (1988)Spuirean na h-Iolaire (1989)
An Neamhnaid Luachmhor (1990)Bàrdachd an Roinn-Eòrpa (1990)
A' Sireadh an Sgadain (1990)Am Fear Meadhanach (1992)
Taigh-Dubh ann an Arnol (1994)Hiort (1995)
Tuath is Tighearna (1995)Duanaire Colach (1997)Moch is Anmoch (1998)
Bàrdachd Dhòmhnaill Alasdair (1999)Smuaintean fo Éiseabhal (2000)
Sgeulachdan Dhòmhnaill Alasdair (2001)Dàin do Eimhir (2002)
Tocasaid 'Ain Tuirc (2004)Na Klondykers (2005)
Saoghal Bana-mharaiche (2007)
ceud 1st
ceud 100
ciad 1st
ciad 100
121
Figure 6-6 Frequency of beul/bial in 18th-century corpus texts
The occurrences in Ais-éiridh (1751) are the oldest instances of bial in the corpus.
This text has 5 beul and 3 bial. As noted in the previous section, MacMhaighstir
Alasdair spoke a central dialect. The use of dialectal forms for rhymes in
MacKinnon's paradigm is illustrated here, for example, in this verse where he
rhymes bial with fial, triall, and cliamh:
Na tostachan siggeanta fial,'G an aiseag go rige mo bhial;Ba mhireagach stuiggidh, is triall,Am marsal le ciogguilt roi' m' chliamh (Mac-Dhonuill 1751: 148)
He also uses beul for rhyme as in the next example where it rhymes with fein,
téud and ré:
Bha smeorach cur na smúid dhith,Air baccan cuil lea fein;An dreadhunn-donn go súrdoil,’S a rifeid chiuil na bheul;Am briccein-beith’ is lúb air,’S e ’g gleusadh lú a théud;An coileach-dubh ri dúrdan;’S a chearc ri túchan ré. (Mac-Dhonuill 1751: 83)
We can see that MacKinnon's paradigm of acceptable variation (in poetry and
dialectal material) is therefore in action from the start of Modern Scottish Gaelic
orthography.
In Figure 6.2, the continued dominance of beul through the nineteenth century is
evident.
0 10 20 30 40 50 60 70 80
Gairm an De Mhoir (1750)Ais-éiridh (1751)
Tiomnadh Nuadh (1767)Leabhar Ceasnuighe Aithleasuighte (1779)
Saighdear Criosduidh (1797)
beul
bial
122
Figure 6-7 Frequency of beul/bial in 19th-century corpus texts
The occurrences of bial are mainly in poetic contexts although even in these
contexts, beul is the norm, i.e. in Sar-Obair nam Bard Gaelach (1841), Laoidhean
agus Dàin (1868), An t-Oranaiche (1879), Dàin agus Orain Ghàidhlig (1891), Dàin
Iain Gobha 1 (1893), Na Bàird Leathanach (1898) and Leabhar na Ceilidh (1898).
While Orain ann sa Ghailig (1875) looks as though it uses only <ia>, there is only
one instance of bial where it is used to rhyme with diadhachd, blianach, sian.
Including the other lemmas, Orain ann sa Ghailig (1875) has 3 <eu> occurrences
and 3 <ia> occurrences.
Ordo Missae (1877) also has only one instance of bial but also only uses <ia> in
ciad (2 occurrences as 'first', 1 as 'hundred') and it is the earliest text in the corpus
that does not fall within MacKinnon's acceptable variation paradigm. The text is
the Catholic Order of Mass in Gaelic. While its dialect origin is unknown, the use
of <ia> suggests a central dialect area source. Its nature as a Catholic text is also
possibly significant. Other writers in the later nineteenth century who promote
the Bible as a standard are following Protestant precedent and the rulings of the
Church of Scotland on approved versions. A Catholic translator is perhaps less
123
likely to be inclined to consider Protestant texts as the last word on the matter
and be more open to non-standard uses of <ia>. Similarly, Lòchran an Anma (1906)
which only uses the <ia> forms of bial (6 occurences) and ciad (3 occurences) is a
book of Catholic prayers.
Between 1900 and 1980 (see Figure 6.3) the frequency of bial increases with five
texts using bial only and another two using bial more often than beul; Folksongs
and Folklore (1955) and Sporan Dhòmhnaill (1968). Of the 51 texts in the corpus
that date from 1900 to 1980 inclusive, 25 of them use bial. Of those, 15 texts are
poetry and folkloric texts. However, 10 of them are not within this paradigm but
are more clearly challenging <eu> as a norm. They include:
An t-Ogha Mor (1913) fictionLòchran an Anma (1906) religious proseLitrichean Alasdair Mhoir (1932) newspaper columnsBriseadh na Cloiche (1970) fictionLuach na Saorsa (1970) essay and autobiographyAn t-Aonaran (1976) fictionTìr an Aigh (1971) fictionAitealan Dlù is Cian (1972) essaysNach Neònach Sin (1973) fictionOighreachd is Gabhaltas (1980) academic essays
The range of literary contexts in these works move beyond MacKinnon's paradigm
of acceptable variation. It is significant that most of these are from the 1970s and
therefore in the period immediately preceding the first GOC report. This pattern
will be repeated in the data of the other <eu>/<ia> lemmas.
125
After the first GOC report in 1981, bial recedes and reverts to MacKinnon's
paradigm, with bial appearing only in Bàrdachd na Roinn-Eòrpa (1990), Duanaire
Colach (1997), Smuaintean fo Éiseabhal (2000) and Dàin do Eimhir (2002).
Figure 6-9 Frequency of beul/bial between 1981 and 2005
The most recent example is in Christopher Whyte's 2002 edition of Sorley
MacLean's Dàin do Eimhir. MacLean's dialect of Raasay is within the central dialect
area. Whyte is explicit about his editing decisions and specifies that while he has
modernised the spelling generally he kept:
the careful distinction made in MacLean's later versions of his textsbetween diphthong ('bial') and monophthong ('eu') pronunciations of thesame word (MacLean 2002: 124)
6.3.4 sgeul vs sgial
The difference between sgeul and sgial is only one of dialectal pronunciation..
They both have the meaning of 'story'. There are 1,224 occurrences of sgeul in the
126
corpus and 104 of sgial. Although there are no occurrences of sgial in the
eighteenth-century corpus texts, the oldest reference to sgial in the corpus does
come from the eighteenth century and appears in Ó Baoill's edition of MacLean
poetry, Eachann Bacach (1979). In reference to the poem 'Gabhaidh mi Sgeula
dem' Shagart' by Iain Mac Ailein, Ó Baoill notes a variant spelling in the MS made
by Dr Hector Maclean of Grulin, Eigg, who died about 1785 (Ó Baoill 1979: 310).
The MS is believed to have been in progress from around 1738 until Maclean's death
(Ó Baoill 1979: 43).28 Eigg falls in the central dialect area where SGDS (Point 94)
records broken e.
Figure 6-10 Frequency of sgeul/sgial in 18th-century corpus texts
As Figure 6.6 below shows, sgial is also rare in the nineteenth century. Although
there is one early occurrence in 1835, it does not reappear until the last decade
of the century albeit appearing only once per text and in books of poetry. In its
appearance in An Teachdaire Ùr Gàidhealach (1835-6), an early Gaelic periodical,
it appears in the folk tale entitled 'Sean Sgeul' [Old Tale] by an author credited
only as Mac 'Ic Eachainn, Airdghobhar (MacEachen, Ardgour). Although the author
is not identified beyond the surname (or patronymic) MacEachen, the district of
Ardgour falls within the central dialect area where the SGDS data for Ardgour
(Point 78) also recorded broken é. The story also contains occurrences of ciadabh,
bial, sia, diag and ciadna confirming the extensive spread of the diphthong in this
area in the early nineteenth century.
28 Elsewhere, Ó Baoill specifies the unusual spelling of the 'HM' MS: "Tha litreachadh annasachann an HM … Is ann an Gàidhlig dhùchasach na h-Albann a rinn Eachann an dàn, tha mismaoineachadh, ach tha àiteachan ann cuideachd far an tug e cruthan fhacal on GhàidhligChlasaicich (Ó Baoill 1997: 49). (There is unusual spelling in HM ... It was in Scottish Gaelicthat Hector wrote the poem, I believe, but there are also occasions where he took word formsfrom Classical Gaelic)
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
Gairm an De Mhoir (1750)Ais-éiridh (1751)
Tiomnadh Nuadh (1767)Leabhar Ceasnuighe Aithleasuighte (1779)
Saighdear Criosduidh (1797)
sgeul
sgial
127
Figure 6-11 Frequency of sgeul/sgial in 19th-century corpus texts
As with bial, the use of sgial grows in the twentieth century, although there are
only three texts that use it exclusively; An t-Ogha Mor [The Great Grandchild]
(1913), Saoghal an Treobhaiche [The World of the Ploughman] (1972), and
Folksongs and Folklore (1955) (with only 1 occurrence). In the novel An t-Ogha
Mor (1913) the author, Aonghas MacDhonnachaidh (Angus Robertson) made an
explicit choice to use his dialect of Skye in his prose and specifically <ia> over
<eu> (see preface 1913: v). For all the lemmas analysed here, An t-Ogha Mor uses
only the <ia> forms. Saoghal an Treobhaiche (1972) with 17 occurrences, is also
prose, however it can be classified as dialectal material from Uist as the editor
(and transcriber) John Lorne Campbell's interest is in reproducing modern Gaelic
vernacular speech and that is reflected in his orthographic choices. 29 In the
twentieth century overall, then, the variation largely fits MacKinnon's paradigm.
29 Campbell proudly describes the book as 'by far the longest text in modern colloquial ScottishGaelic that we possess.' (1972: 2)
0 10 20 30 40 50 60 70 80 90
Orain Ghaelach (1801)Searmona (1804)
Comhchruinneacha (1813)Co'chruinneachadh (1828)
Daoine air an Comhairleachadh (1832)An Teachdaire Ur Gàidhealach (1835-6)
Sar-Obair nam Bard Gaelach (1841)Laoidhean Spioradail (1862)Laoidhean agus Dain (1868)
Am Filidh Gaidhealach (1873)Orain ann sa Ghailig (1875)
An t-Oranaiche (1879)Gaelic Songs (1880)
Poems and Songs (1880)Poems (1884)
Marbh-rainn ... (1887)Dain agus Orain Ghaidhlig (1891)
Dàin Iain Ghobha 1 (1893)Dàin Iain Ghobha 2 (1896)
Na Bàird Leathanach (1898)Leabhar na Ceilidh (1898)
sgeul
sgial
128
Although sgial appears in Ugam Agus Bhuam (1977), it is only on one occasion,
where the editor states that it is a form he will not be using.
129
Figure 6-12 Frequency of sgeul/sgial between 1900 and 1980
After the first GOC reforms, sgial appears only once more in the corpus.
0 10 20 30 40 50
Laoidhean Bean Torra Dhamh (1906)Oiteagan o'n Iar (1908)
Am Fear-Ciuil (1910)Aig Tigh Na Beinne (1911)
Dùn-Aluinn (1912)An t-Ogha Mor (1913)
A' Bhraisd Lathurnach (1914)Cailin Sgiathanach (1923)
Na Baird Thirisdeach (1932)Litrichean Alasdair Mhoir (1932)
Highland Songs of the Forty-Five (1933)Gaelic Songs of Mary MacLeod (1934)
Baird Chill-Chomain (1936)Clarsach nam Beann (1937)
Am Measg nam Bodach (1938)Songs of John MacCodrum (1938)
Ban-altrumachd aig an Tigh (1939)Griasaiche Bhearnaraidh (1940)
Sgialachdan Dhunnchaidh (1944)Folksongs and Folklore (1955)
Orain Iain Luim (1964)Sporan Dhòmhnaill (1968)
HebrideanFolksongs 1 (1969)Dòmhnall Ruadh Chorùna (1969)
Briseadh na Cloiche (1970)Luach na Saorsa (1970)An Clarsair Dall (1970)
Tìr an Aigh (1971)Lus-Chrun a Griomasaigh (1971)Saoghal an Treobhaiche (1972)
Aitealan Dlù is Cian (1972)A' Bhratach Dhealrach (1972)
Suathadh ri Iomadh Rubha (1973)Nach Neònach Sin (1973)
Criomagan Ioma-dhathte (1973)Creach Mhor nam Fiadh (1973)
Fo Sgail a' Swastika (1974)An t-Aonaran (1976)
Bith-eòlas (1976)Ugam agus Bhuam (1977)
Hebridean Folksongs 2 (1977)Orain Dhonnchaidh Bhain (1978)
Eachann Bacach (1979)Deireadh an Fhoghair (1979)
Oighreachd is Gabhaltas (1980)
sgeul
sgial
130
Figure 6-13 Frequency of sgeul/sgial post-GOC
The last appearance of sgial is in the poetry collection Tuath is Tighearna (1995)
(7 sgeul, 1 sgial). Again, the example fits MacKinnon's paradigm as it is in a
nineteenth-century poem, 'Oran Muinntir Bheàrnaraidh' by Murchadh MacLeòid
(87, line 54) and is used for rhyme.
6.3.5 geur vs giar
In total there are 881 occurrences of geur and only 21 of giar. Both have the
meaning of 'sharp'. There are no occurrences of giar in the eighteenth century
texts.
Figure 6-14 Frequency of geur/giar in 18th-century corpus texts.
0 10 20 30 40 50 60 70 80
Hebridean Folksongs 3 (1981)Air Mo Chuairt (1982)
Creachadh na Clàrsaich (1982)Suileabhan (1983)
Air Druim an Eich Sgiathaich (1987)Sgrìobhaidhean Choinnich MhicLeòid…
Spuirean na h-Iolaire (1989)An Neamhnaid Luachmhor (1990)
Bàrdachd an Roinn Eòrpa (1990)Am Fear Meadhanach (1992)
Hiort (1995)Tuath is Tighearna (1995)
Duanaire Colach (1997)Moch is Anmoch (1998)
Bàrdachd Dhòmhnaill Alasdair (1999)Smuaintean fo Eiseabhal (2000)
Sgeulachdan Dhòmhnaill Alasdair (2001)Dàin do Eimhir (2002)
Tocasaid 'Ain Tuirc (2004)Na Klondykers (2005)
Saoghal Bana-mharaiche (2007)
sgeul
sgial
0 5 10 15 20 25 30
Ais-éiridh (1751)Tiomnadh Nuadh (1767)
Leabhar Ceasnuighe Aithleasuighte (1779)Saighdear Criosduidh (1797)
geur
giar
131
As with sgial, giar does not appear in the corpus until An Teachdaire Ùr
Gàidhealach (1835-6) (geur: 10, giar: 1). The single occurrence appears in the
unattributed poem entitled 'Trasg' where it is used to rhyme with biadh:
Cha 'n è! 'S e trasg do choirce thabhairtDo 'n anam acrach, sguab a's biadh;'Bhi trasg o chonnsach', fuath a's gamhlas,Do chridhe nàimhdeil 'cheusadh giar; (1835-6: 29)
The same poem also uses beul to rhyme with céin, demonstrating the use poets
made of variation in pronunciation.
Giar remains rare in the nineteenth century, appearing only in poetic contexts in
the texts Orain ann sa Ghailig (1875), An t-Oranaiche (1879), Dàin Iain Ghobha 2
(1896) and Na Bàird Leathanach (1898) where in each example it is used for rhyme.
Figure 6-15 Frequency of geur/giar in 19th-century corpus texts
As Figure 6.11 shows, in the twentieth century, Lòchran an Anma (1906) and An t-
Ogha Mor (1913), both of which only use <ia> forms, are the only two texts to use
<ia> exclusively with giar.
0 10 20 30 40 50 60 70 80 90
Orain Ghaelach (1801)Searmona (1804)
Comhchruinneacha (1813)Co'chruinneachadh (1828)
An Teachdaire Ur Gaidhealach (1835-6)Beachd-Chomhairlean (1838)
Sar-Obair nam Bard Gaelach (1841)Laoidhean agus Dain (1868)
Am Filidh Gaidhealach (1873)Orain ann sa Ghailig (1875)
An t-Oranaiche (1879)Poems and Songs (1880)
Laithean Ceisde (1880)Dan Spioradail (1885)
Marbh-rainn ... (1887)Dain agus Orain Ghaidhlig (1891)
Dàin Iain Ghobha 1 (1893)An t-Urramach Iain Mac-Rath (1895)
Dàin Iain Ghobha 2 (1896)Na Baird Leathanach (1898)
Leabhar na Ceilidh (1898)
geur
giar
132
Figure 6-16 Frequency of geur/giar between 1900 and 1980
The one appearance of giar in Eachann Bacach (1979) is in 'Moladh na Pìoba' by
Iain Mac Ailein of Mull (c. 1650 – c. 1738). In this case it is not an archaic spelling,
the MS originals having geur (Ó Baoill 1979: 305), but appears to be part of Ó
0 5 10 15 20 25
Laoidhean Bean Torra Dhamh (1906)Lochran an Anma (1906)
Oiteagan o'n Iar (1908)Am Fear-Ciuil (1910)
Aig Tigh Na Beinne (1911)Dùn-Aluinn (1912)
An t-Ogha Mor (1913)Dugald Buchanan (1913)
Cailin Sgiathanach (1923)Na Baird Thirisdeach (1932)
Litrichean Alasdair Mhoir (1932)Highland Songs of the Forty-Five (1933)
Baird Chill-Chomain (1936)Clarsach nam Beann (1937)
Am Measg nam Bodach (1938)Songs of John MacCodrum (1938)
Ban-altrumachd aig an Tigh (1939)Griasaiche Bhearnaraidh (1940)
Sgialachdan Dhunnchaidh (1944)Folksongs and Folklore (1955)
Bùrn is Aran (1960)An Dubh is An Gorm (1963)
Orain Iain Luim (1964)Sporan Dhòmhnaill (1968)
Dòmhnall Ruadh Chorùna (1969)Briseadh na Cloiche (1970)
Luach na Saorsa (1970)An Clarsair Dall (1970)
Tìr an Aigh (1971)Lus-Chrun a Griomasaigh (1971)Saoghal an Treobhaiche (1972)
Aitealan Dlù is Cian (1972)A' Bhratach Dhealrach (1972)
Bardachd Shilis na Ceapaich (1972)Suathadh ri Iomadh Rubha (1973)
Nach Neònach Sin (1973)Criomagan Ioma-dhathte (1973)An Aghaidh Choimheach (1973)Creach Mhor nam Fiadh (1973)
Fo Sgail a' Swastika (1974)Gàidhlig ann an Albainn (1976)
An t-Aonaran (1976)Ugam agus Bhuam (1977)
Hebridean Folksongs 2 (1977)Orain Dhonnchaidh Bhain (1978)
Eachann Bacach (1979)Deireadh an Fhoghair (1979)
geur
giar
133
Baoill's modernisation. It is used to rhyme with cian in the quote below, and with
bial and miar further on:
'S dearbha gu robh e stuidearra trom'S a shusbainte giar,'M fear a smuaintich an toiseach gun coisneadh i bonn'S fortan do chian (Ó Baoill 1979: 303)
Ó Baoill modernises the orthography generally in this edition of poetry.30
After GOC, giar falls out of use and only appears, as with the example of bial, in
MacLean's collection of poetry Dàin do Eimhir (2002) (geur: 6, giar: 2).
Figure 6-17 Frequency of geur/giar post-GOC
Compared with bial and sgial, giar is more limited to MacKinnon's paradigm.
Several of the texts in the 1970s which use bial and sgial do not extend <ia> to
<giar>; for example Tìr an Aigh (1971) (Skye), Saoghal an Treobhaiche (1972)
(Uist), Aitealan Dlù is Cian (1972) (Skye) and Nach Neònach Sin (1973) (Harris).
30 '…I have attempted to present in an acceptable modern spelling. This means rejecting localdialect forms and archaic forms (like chuaidh, go bhfuil) present in the sources, except wherethey are clearly part of the rhyme of metrical system' (Ó Baoill 1979: viii)
0 5 10 15 20 25
Hebridean Folksongs 3 (1981)Air Mo Chuairt (1982)
Creachadh na Clàrsaich (1982)Suileabhan (1983)
Air Druim an Eich Sgiathaich (1987)Sgrìobhaidhean Choinnich MhicLeòid…
Spuirean na h-Iolaire (1989)An Neamhnaid Luachmhor (1990)
Bàrdachd an Roinn Eòrpa (1990)A' Sireadh an Sgadain (1990)
Iasgach (1991)Am Fear Meadhanach (1992)
Hiort (1995)Tuath is Tighearna (1995)
Duanaire Colach (1997)Moch is Anmoch (1998)
Bàrdachd Dhòmhnaill Alasdair (1999)Sgeulachdan Dhòmhnaill Alasdair (2001)
Dàin do Eimhir (2002)Tocasaid 'Ain Tuirc (2004)
Na Klondykers (2005)Saoghal Bana-mharaiche (2007)
geur
giar
134
6.3.6 feur vs fiar
In total, there are 327 occurrences of feur and 147 of fiar. While feur has the
primary meaning of 'grass',31 fiar can be both 'grass' and the adjective 'crooked,
curved, bent', or the verb 'to bend, twist, make crooked'.32 In the first GOC report,
the historical differentiation between feur and fiar was re-established:
When there are homophones, however, these should be differentiatedin spelling where such differentiation is feasible. (...) For example (...)feur = "grass"; fìor = "true"; fiar = "squint". (SEB 1981: 4)
This was repeated in GOC 2005 and 2009. Of the occurrences of fiar in the corpus,
62 are with the meaning 'squint'. This leaves 327 occurrences of feur, 'grass' and
72 of fiar, 'grass'.
Figure 6-18 Frequency of feur/fiar 'grass' in 18th-century corpus texts
There are no occurrences of fiar, 'grass' in the eighteenth century. Figure 6.14
below shows that the earliest occurrence of fiar, 'grass', is in Sar-Obair nam Bard
Gaelach (1841) where feur is more common (feur: 25, fiar: 2) where it rhymes
with sliabh in 'Cuachag nan Craobh' by Uilleam Ros:
Gur gile do bhian na sneachd air an fhiar,'S na canach air sliabh mointich (MacKenzie 1841: 293)
Ros was from Skye and the editor, MacKenzie, from Gairloch, both central dialect
areas. The other example is from 'Oran air Blar na h-Eiphit' by Alasdair Mac-
Iomhnuinn <sic> [Alexander MacKinnon] (MacKenzie 1841: 344). MacKinnon (1770-
1814) was from Morar, another central dialect (MacKenzie 1841: 339-40). As with
the previous lemmas, the nineteenth-century occurrences appear in poetical
works, Sar-Obair nam Bard Gaelach (1841), Am Filidh Gaidhealach (1873), Dàin
agus Orain Ghàidhlig (1891) and Dàin Iain Ghobha 2 (1896). In fact the single
occurrence in Am Filidh Gaidhealach (1873) is in the same line of poetry by
31 'Grass. 2 Herbage. 3 Hay'. (Dwelly 1911: s.v. feur).32 'Bend, twist, make crooked. 2 Pervert. 3 Go astray or aside'. (Dwelly 1911: s.v. fiar)
0 2 4 6 8 10
Ais-éiridh (1751)Tiomnadh Nuadh (1767)
feur
fiar
135
Uilleam Ros that appeared in Sar-Obair nam Bard Gaelach (1841). Both Dàin agus
Orain Ghàidhlig (1891) and Dàin Iain Ghobha 2 (1896) are collections of poems
from central dialect areas, Skye and Harris respectively.
Figure 6-19 Frequency of feur/fiar 'grass' in 19th-century corpus texts
The twentieth century sees an increase in the occurrence of fiar, similar to the
previous word pairs. Many of the texts previously noted as using <ia> are found
again with this lemma; i.e. Lòchran an Anma (1906), Litrichean Alasdair Mhoir
(1932), Sgialachdan Dhunnchaidh (1944). The particularly high frequency of fiar
in Saoghal an Treobhaiche is because it is the autobiographical account of a farm
labourer, a domain in which grass and hay are more frequently referred to.
The most recent occurrence of fiar is in Hebridean Folksongs 2 (1977) where is it
the more common form (feur: 1, fiar: 4). The texts that move outside MacKinnon's
paradigm are Litrichean Alasdair Mhoir (1932), which is a collection of light-
hearted newspaper columns, and the modern literary works of Briseadh na Cloiche
(1970), Aitealan Dlù is Cian (1972) and Nach Neònach Sin (1973).
0 5 10 15 20 25 30
Orain Ghaelach (1801)Searmona (1804)
Comhchruinneacha (1813)Co'chruinneachadh (1828)
An Teachdaire Ur Gaidhealach (1835-6)Beachd-Chomhairlean (1838)
Sar-Obair nam Bard Gaelach (1841)Laoidhean agus Dain (1868)
Am Filidh Gaidhealach (1873)Orain ann sa Ghailig (1875)
An t-Oranaiche (1879)Poems and Songs (1880)
Poems (1884)Croft Cultivation (1885)
Dain agus Orain Ghaidhlig (1891)Dàin Iain Ghobha 1 (1893)
An t-Urramach Iain Mac-Rath (1895)Dàin Iain Ghobha 2 (1896)
Na Baird Leathanach (1898)Leabhar na Ceilidh (1898)
feur
fiar
136
Figure 6-20 Frequency of feur/fiar 'grass' in 20th-century texts pre-GOC
As Figure 6.16 below shows, there are no uses of fiar with the meaning 'grass' in
the corpus texts after GOC 1981. Within the corpus texts then, the
recommendation in GOC to remove homographs and to use the <eu>/<ia> variation
to distinguish between, feur, 'grass' and fiar, 'squint', has been followed. While
there are 7 occurrences of fiar in the corpus post-GOC, they all have the meaning
of 'squint'.
0 5 10 15 20 25 30 35
Lochran an Anma (1906)Oiteagan o'n Iar (1908)
Am Fear-Ciuil (1910)Aig Tigh Na Beinne (1911)
Dùn-Aluinn (1912)Spiritual Songs of Dugald Buchanan (1913)
Cailin Sgiathanach (1923)Na Baird Thirisdeach (1932)
Litrichean Alasdair Mhoir (1932)Highland Songs of the Forty-Five (1933)
Gaelic Songs of Mary MacLeod (1934)Clarsach nam Beann (1937)
Am Measg nam Bodach (1938)Sgialachdan Dhunnchaidh (1944)
Folksongs and Folklore (1955)Bùrn is Aran (1960)
An Dubh is An Gorm (1963)Orain Iain Luim (1964)
Sporan Dhòmhnaill (1968)HebrideanFolksongs 1 (1969)
Dòmhnall Ruadh Chorùna (1969)Briseadh na Cloiche (1970)
Luach na Saorsa (1970)Tìr an Aigh (1971)
Lus-Chrun a Griomasaigh (1971)Saoghal an Treobhaiche (1972)
Silis na Ceapaich (1972)Aitealan Dlù is Cian (1972)
A' Bhratach Dhealrach (1972)Suathadh ri Iomadh Rubha (1973)
Nach Neònach Sin (1973)Criomagan Ioma-dhathte (1973)An Aghaidh Choimheach (1973)
Fo Sgail a' Swastika (1974)Bith-eòlas (1976)
Ugam agus Bhuam (1977)Hebridean Folksongs 2 (1977)
Orain Dhonnchaidh Bhain (1978)Eachann Bacach (1979)
Deireadh an Fhoghair (1979)
feur
fiar
137
Figure 6-21 Frequency of feur/fiar, 'grass', post-GOC
6.4 Summary and Conclusions
The corpus data which shows variation between <eu> and <ia> illustrates that
Gaelic spelling usage previously allowed for a level of variation within the
standard. The corpus evidence supports the central and peripheral dialectal
distinction as <ia> occurrences only appear in central dialect texts or texts where
the dialect is unknown. The use of <ia> was also largely restricted to what I have
termed MacKinnon's paradigm, except in the case of ciad. The nineteenth century
saw a growth of texts in these contexts being produced, spurred on by the interest
in folklore following John F. Campbell's Popular Tales of the West Highlands (1860)
and the creation of Gaelic associations such as An Comunn Gàidhealach who were
interested in preserving and promulgating the oral culture of the Highlands. These
are the texts where the growth in the use of <ia> in contrast to <eu> can be seen.
Uniformity is a key feature of the ideology of the standard, however, and in the
nineteenth and early twentieth century variation was only welcome in certain
contexts. This is seen in MacKinnon's paradigm.
There was also a period where the value of dialects pushed back against <eu> as
the standard form. In the 1970s particularly, there was a trend for central dialects
0 5 10 15 20 25 30 35
Hebridean Folksongs 3 (1981)Creachadh na Clàrsaich (1982)
Suileabhan (1983)Seann Taighean Tirisdeach (1986)
Eòin an Àite (1986)Loch Druidibeg (1988)
Sgrìobhaidhean Choinnich MhicLeòid…Spuirean na h-Iolaire (1989)
Bàrdachd an Roinn Eòrpa (1990)Croitearachd (1991)
Am Fear Meadhanach (1992)Taigh-Dubh ann an Arnol (1994)
Hiort (1995)Tuath is Tighearna (1995)
Duanaire Colach (1997)Saoghal Bana-mharaiche (2007)
Bàrdachd Dhòmhnaill Alasdair (1999)Smuaintean fo Eiseabhal (2000)
Tocasaid 'Ain Tuirc (2004)Na Klondykers (2005)
feur
fiar
138
to use <ia> to bring the spelling of certain lemmas closer to the phonology of their
speech. This happened in the wider linguistic context of a diminishing population
of speakers of peripheral dialects both in real terms and as a percentage of Gaelic
speakers. The use of <ia> expanded beyond the 'acceptable variation' use in poetry
and dialectal materials to using it in prose writing. By the 1970s, the variation was
spreading outwith MacKinnon's paradigm into literature more generally. This can
be illustrated by the texts of Iain Mac a' Ghobhainn (Iain Crichton Smith). Two of
his works which were published in the 1960s, Bùrn is Aran (1960) and An Dubh
agus an Gorm (1963), contain none of the <ia> lemmas analysed here. When his
short novel An t-Aonaran was published in 1976, however, he uses bial 11 times
with no uses of beul.
This attempted normalisation of a marked dialectal form had the potential to
replace the existing <eu> form or to establish itself as an equal variant, not
dependent on 'appropriate' contexts. In Chapter 2.2.1.3, the concept of polynomie
and dialect variation as source of sociolinguistic value was discussed. Given the
value placed on dialects as seats of language authority in the Gaelic community,
the expansion of <ia> forms in the 1970s can be understood as a consequence of
this. Given the value of dialects, it could have been possible to have 'polynomic'
spellings. This is to some extent endorsed by GOC 1981, except for where it
advocated using <eu> and <ia> to differentiate between homophones as with
feur/fiar. The notion of polynomie for minority languages is complicated,
however, by the equally accepted notions of 'right' and 'wrong' in language (the
ideology of the standard) that has been central to Western education. In the case
of the <ia> variant, this expansion of <ia> was rapidly reversed in the 1980s and
there was the entrenchment of the conservative <eu> practice. While in Dwelly
(1911) the headword entries for bial, etc. refer readers to the <eu> form, in Colin
Mark's Gaelic-English Dictionary (2003), <ia> forms are not listed at all. Mark might
have pre-empted GOC 2005 and 2009's removal of <ia> but is, in fact, following
the usage that has preferred only one form in the orthography. The current online
dictionary at LearnGaelic.net (a partnership between the official bodies of the
BBC, Sabhal Mòr Ostaig, the Board of Celtic Studies and Bòrd na Gàidhlig)
automatically redirects bial to beul, sgial to sgeulachd, nial to neul and friamh to
freumh.
139
The pressures for continuity with established spelling, the concerns in education
over poor literacy standards and the expectation of all stakeholders (teachers,
parents and pupils) for there to be clear 'rights' and 'wrongs' to be taught has
meant the removal of <ia> variants, despite the phonographic impetus that would
bring those spellings closer to the majority of contemporary dialects. To return to
Coulmas' notion of real writing systems being compromise, historic and pragmatic,
we can see that the <eu>/<ia> variation and its standardisation in Gaelic spelling
can be interpreted in these terms. There was compromise in that <ia> was, and
continues to occasionally be, a variation appropriate to poetry and a conventional
representation of dialectal /ia/ (a 'marked' norm). It is historic in that the
conservative form, based on dialects that receded during the twentieth century,
has been maintained as the unmarked standard form. It is pragmatic in that the
opportunity to remove an ambiguity, that between feur, 'grass'/fiar, 'squint', has
been adhered to.
140
7 Uniformity vs Continuity: <st>, <sp> and <sg>
7.1 Introduction
In the Rules for Reading the Gailic Language prefacing his 1795 dictionary, Robert
Macfarlan drew attention to an interchangeability in the use of these consonant
clusters:
It may be further noticed, that sb, sp, sd, st, sc and sg, are sometimeswritten the one for the other, or used indifferently; but as b, d, and g,when joined to an s, have a softer sound than the other, care should betaken, when to use the one and when the other. (Macfarlan 1795: 18)
His comments on usage reveal both the real writing practice of variable usage and
the belief in an improved system where different spellings should serve particular
purposes.
In Scottish Gaelic writing, there are three pairs where <s> is followed by stop
consonants: <sb>/<sp>, <sg>/<sc> and <sd>/<st>. Each cluster has, however, only
one phonetic realisation, respectively /sp/, /sk/ and /st/, as the opposition
between the voiceless and voiceless aspirated consonants is neutralized after
/s/.33 Any variation in spelling, therefore, is not due to any dialectal variation or
pronunciation change but is an orthographic matter. Regarding loan words,
however, it is a feature influenced to some extent by the spelling in the language
of origin. All six of the clusters have been used in all word positions at different
points in the language's history. Variation between each pair can be seen in the
Old and Middle Gaelic citations that form the basis of the Dictionary of the Irish
Language (DIL) produced by the Royal Irish Academy. The early printed works in
Scottish Gaelic, Foirm na n-Urrnuidheadh (1567) and Adtimchiol an Chreidimh
(c.1630), also demonstrate varied and inconsistent usage.
The corpus analysis in this chapter shows how the usage of these clusters became
standardised and more consistent. This process of standardisation first reduced
the clusters to only <sp>, <st>, <sg> word initially. Then the use of the other pairs
33 Macfarlan's comments suggest a distinction that some native speakers have also claimed toperceive. However, it is possible that their perception is influenced by their knowledge ofspelling: a form of spelling pronunciation. An investigation into perceived and phonologicaldistinctions between the consonant groups is beyond the scope of this study.
141
was standardised in medial and final position, with usage being reformed and
codified in the GOC reports of the late twentieth century to <st> only (except in
proper nouns), <sg> only, and <sp> with some exceptions.
The first GOC report covered only <st>/<sd> (SEB 1981: 2.2). Details from all three
reports are given in the relevant sections below. The decisions in the GOC reports
to reduce and simplify the usage of these clusters, particularly <sd> and <st>,
demonstrates the increased desire for consistency, regularisation and simplicity
in the second half of the twentieth century.
7.2 Corpus Analysis of s + stop consonants clusters
WordSmith Tools was first used to retrieve the consonant clusters <sb>, <sp>, <sc>,
<sg>, <st> and <sd> in any word position.
Some instances were returned where <'s>, a form of both the copula, is, or the
reduced form of agus, 'and', was printed attached to the initial letter of the
following word as in the following table. While this feature raises interesting
questions regarding the representation of word boundaries, it is not relevant to
this investigation and so these instances were discarded from the analysis.
Table 7-1 Examples of is attached to following consonantOccurrence Standard spelling Text'scha mho 's cha mho Beachd-Chomhairlean (1838)scòir dhuibh 's còir dhuibh Searmona (1804)'sdòcha 's dòcha Bàrdachd na Roinn-Eòrpa (1990): Bàrdachd Neruda'sdù' 's dùth An Teachdaire Ùr Gàidhealach (1835-6)sgun 's gun Co'chruinneachadh (1828)sdu Is tu (?) Duanaire Colach (1997)
Other instances were retrieved where the process of compounding placed <s>
before <c>, <g>, <p> or <b>. These results were removed before the analysis as
these examples do not reflect the same phonological context as the stop
consonant clusters investigated here. For these, the more standard form (as shown
in the headword in Dwelly) would be to separate the two elements with a hyphen
or with a space in the case of os cionn. Table 7.2 below shows these instances.
142
Table 7-2 Compounds with medial <sc>, <sg> and <sb>Lemma Dwelly
headwordText No. of
occurr.osbar os barr Leabhar Ceasnuighe Aithleasuighte
(1779)Comhchruinneacha (1813)An Teachdaire Ùr Gàidhealach (1835-6)Laoidhean agus Dàin (1868)
4
oscionn /osceann
os cionn Saighidear Criosduidh (1797) 1
Searmona (1804) 19
An Teachdaire Ùr Gàidhealach (1835-6) 6
Beachd-Chomhairlean (1838) 3
Sar-Obair nam Bard Gaelach (1841) 3
casbhuidhe cas-bhuidhe Comhchruinneacha (1813)Orain Dhonnchaidh Bhàin (1978)
2
fearaschuideachd fearas-cuideachd
Na Bàird Leathanach (1898) 1
basbhualadh bas-bhualadh Gaelic Songs of Mary MacLeod (1934)Songs of John MacCodrum (1938)Na Klondykers (2005)
3
cìoschain cìos-chàin Orain Iain Luim (1964) 1glascharn glas-chàrn Orain Iain Luim (1964) 1claisghorm clais ghorm Orain Iain Luim (1964) 1cascheann cas-cheann Orain Iain Luim (1964) 1sliosbhord n/a Orain Iain Luim (1964) 1cluasbhiorach cluas-bhiorach Orain Dhonnchaidh Bhàin (1978) 2cascheum cas-cheum Smuaintean fo Éiseabhal (2000) 1fàschoill fàs-choille Smuaintean fo Éiseabhal (2000) 1
Some other words were returned that have clusters of <sbh> that do not represent
/sp/ and are not conventionally separated. Although the GOC reports of 2005 and
2009, if taken literally, would replace <sb> in all positions, it is unlikely that they
would be interpreted to apply in this situation where the cluster is <s+bh> not
<sb>.34
Table 7-3 Compounds with <sbh>Lemma Freq. No. of texts Dates Glossasbhuain 12 10 1835-1982 stubble, 'out-reaping'easbhaidh 182 45 1779-2007 lack, defectuireasbhaidh 137 50 1804-2001 want, povertycaisbheart 15 8 1893-1995 footwear
Proper nouns were not included in this analysis as these are considered exceptions
by the GOC recommendations, e.g. Alasdair, Gilleasbaig (SQA 2005: 3).
34 easbhaidh, easbhaidheach and uireasbhaidh also appear in the 2005 and 2009 GOC wordlists.
143
Assimilated loan words are included where the writer appeared to be using them
as Gaelic lexical items, i.e. if there was any orthographic or morphological
assimilation of the word (other than lenition) e.g. inspeactar, telesgop,
diosteampair. Where compounds such as fior-sgeul appear, they have been
lemmatised and analysed as the element with the relevant consonant cluster e.g.
sgeul.
Each consonant cluster will be considered separately, considering the data word-
initially first, followed by word medially and finally. As some texts would use a
word many times and others only once, the primary analysis here uses the
frequencies of texts, rather than individual occurrences, to avoid skewing the
data.
144
7.3 The <sc> and <sg> clusters
7.3.1 Variability in <sc> and <sg>
The long-standing variability in Old and Middle Gaelic sources in the use of <sc>
and <sg> can be seen in the entries of the Dictionary of the Irish Language (DIL)
(1913-1976). Words with citations using both <sc> and <sg> forms include such
common words as scríbaid (modern sgrìobh, 'write'), scriptuir (modern sgriobtair,
'scripture'), scris (modern sgrios, 'destroy'), scaipid (modern sgap, 'scatter') and
soiscél (modern soisgeul, 'gospel') (see DIL s.v. scríbaid, scriptuir, scris, scaipid,
soiscél).
In his detailed description of the first printed text in Gaelic, Foirm na n-
Urrnuidheadh (1567), R. L. Thomson notes that initial 'sg- is the norm, with only
scāilighe, scaiteach and rarely scrīobhadh, scrīobhtūir having sc-' (Thomson 1970:
xiii). In fact, the more often a word appears, the more likely it is that both
consonant forms will be used. Four words appear ten times or more: of them
sgribtúir, 'scripture', sgandal', 'scandal', and sgrīobh, 'write', all appear with both
<sg> and <sc>, sgris alone (modern sgrios, 'destroy') appears with <sg> only.
Scailighe and scaiteach appear only once each, however following the evidence of
the other lemmas, it seems likely that an <sg> form would be likely if they were
repeated in the text.
Through the seventeenth and eighteenth centuries in Scotland it seems that the
use of <sg> was extended to more widespread and consistent use. Lhuyd's
Dictionary in 1707 lists 103 <sc> headwords and 203 <sg> headwords. By the early
nineteenth century, use of <sc> was the exception rather than the norm. In the
Highland Society Dictionary of 1828, the initial entry for <sc> says 'For most words
beginning with Sc, vide Sg.' and although it lists specifically scairt, scairteach,
scaiteach, scamhan, scòrr, each entry redirects to the <sg> form. MacLeod and
Dewar's 1831 dictionary includes scairt and scor as headwords but both those
entries redirect to their <sg> equivalents as well as noting 'For most words
beginning with Sc, see Sg.' (MacLeod & Dewar s.v. Sc.)
145
Use of <sc>/<sg> was not covered in the 1981 GOC report. In the 2005 and 2009
reports, it was recommended that <sg> be used in all positions and no exceptions
were noted or made in the wordlist.35
We can expect the corpus to demonstrate that <sg> is also the norm. The examples
with <sc> will be considered to see when the shift from <sc> to <sg> is made and
if there are or were any regular exceptions or patterns in the use of <sc>.
7.3.2 Initial <sc> and <sg>
For initial <sg->, the corpus returned 117 lemmas from 20,027 occurrences
whereas for initial <sc->, the corpus only returned 27 lemmas from 197
occurrences. This establishes that <sg-> is clearly the norm word-initially.
The most common lemmas with <sg>-only forms in the corpus are listed in table
7.4 below. The first appearance of each of these lemmas in the corpus is before
1835 and they are not rare in the corpus, each appearing in over 10 texts.
Table 7-4 Words that only appear as <sg->sg- lemma No. of texts Dates Gloss
1 air sgàth 95 1779-2007 for the sake of2 sgadan 57 1801-2007 herring3 sgàil 83 1779-2005 shade4 sgàin 63 1779-2005 burst5 sgainneal 26 1813-2001 scandal, reproach6 sgàirneach 13 1828-1988 scree7 sgal 42 1835-2004 blast of sound8 sgalag 29 1813-2007 skivvy
There were 27 lemmas in the corpus that had examples of initial <sc>. The
following table shows all these <sc> lemmas and their <sg> equivalent forms
together with the dates of appearance and the number of texts they appeared in.
The table is ordered in descending order of frequency of <sg> lemmas.
35 'The letters sg should be used in all positions in place of sc: basgaid, cosg, pasgan, Sgalpaigh,sgian' (SQA 2005: 3).
146
Table 7-5 <sc-> lemmas and their <sg-> equivalentssc- lemma no. of
textsdates sg-
lemmano. oftexts
dates gloss
1 scriobh 2 1797,1804
sgrìobh 234 1779-2007
write
2 scoil 1 1804 sgoil 168 1804-2007
school
3 scap 2 1804,1813
sgap 93 1801-2005
scatter
4 scaoil 1 199736 sgaoil 92 1779-2007
spread out
5 scar 1 1813 sgar 75 1779-2005
separate
6 scillin 1 1804 sgillin 71 1801-2007
shilling
7 scian 1 1832 sgian 64 1813-2007
knife
8 scòrnan 1 1838 sgòrnan 62 1779-2007
throat
9 scleò 1 1801 sgleò 57 1801-2007
mist
10 screuch 1 1828 sgreuch 49 1828-2005
scream
11 scread 1 1813 sgread 36 1813-2004
shriek
12 scàrlaid 8 1804-1972
sgàrlaid 32 1804-2000
scarlet
13 scrioptur 9 1750-1896
sgrioptur 28 1779-2007
scripture
14 sceilp 1 1932 sgeilp 16 1911-2001
shelf
15 scafal 1 1813 sgafal 4 1964-2001
scaffold
16 sceum 1 1932 sgeama 4 1972-2005
scheme
17 scona 1 1973 sgona 3 1992-2005
scone
18 scallach 1 1868 sgallach 2 1841,1938
bald
19 scuch 1 1838 sguch 2 1838,1990
sprain
20 scinn 1 1887 sginn 2 1896,1912
protrude
21 sciallt37 1 1977 sgialt 1 1944 sense22 scorpionaibh 1 1910 sgoirpion 1 1937 scorpion23 scallach 1 1868 sgallach 1 1938 trouble-
some25 sceannach 1 1868 - - - glaring26 scalpan 1 1923 - - - chaff27 score-ig 1 2007 - - - scored
36 From a 1786 text reproduced in Ó Baoill (1997).
147
The corpus shows no lemmas which are regular exceptions to the norm of initial
<sg>. For numbers 1-11 in Table 7.5, it is clear that these are rare uses of <sc> as
the <sg> form in the corpus is far more common. These lemmas also do not appear
after 1838 whereas the <sg> forms are in use up to the twenty first century texts.
Examples 12 and 13, the loan words scàrlaid, 'scarlet', and scriptur, 'scripture' are
the most common initial <sc> lemmas, however, even they are more commonly
found as <sg>.38 Their relative frequency as <sc>, however, indicates that loan
words which have <sc> were more likely to feature <sc> in their Gaelic spelling,
at least when they were first adopted. The other loan words from 'scaffold',
'scheme' and 'scone' also occur more often as <sg> in the corpus. The <sc>
occurrences are again the older forms, showing that modern texts prefer to adhere
to initial <sg>. The only initial <sc> form to be used after the GOC report of 2005
is another loan from 'score', score-ig. This appears in the phrase 'Score-ig e an goal'
in Saoghal Bana-mharaiche (2007) where the editor uses the English spelling with
a Gaelic suffix to make the word transparent to the reader.
Where initial <sc> is found, it is usually earlier than the first use of the initial <sg>
form. As can be seen by the dates, the older initial <sc> form had already become
rare in the nineteenth century and almost disappeared in the twentieth. There is
a chronological development from variation between <sc> and <sg> towards initial
<sg> only. From the corpus texts we can see the final stages of this shift over the
late eighteenth and early nineteenth centuries.
7.3.3 Medial and final <sc> and <sg>
The corpus returned 156 lemmas from 21,183 occurrences of medial or final <sg>
(149 medial, 7 final) and 23 lemmas from 294 occurrences of medial or final <sc>
(18 medial, 5 final).
There are seven lemmas which appear in more than ten texts and only occur in an
<sg> form. They are shown in descending order of frequency in Table 7.6 below.
37 From Pàdruig Moireasdan's Ugam agus Bhuam (1977). Glossed in the text as ciall, explainingthat this is a common form in Uist (1977: 21). Retaining the <c> allows for the link to ciall to bemore apparent.
38 Whether loaned from English scripture or Latin scriptura, the <sc> form would have been familiarto Gaelic writers.
148
Table 7-6 <sg> only lemmas in more than 10 textssg lemma no. of texts dates gloss
1 a-measg 197 1779-2007 among2 cosg 83 1797-2007 spend3 an asgaidh 48 1801-2005 no cost4 casgair 39 1779-2004 slaughter5 crùisgean 26 1813-2004 cruse lamp6 aimlisg 21 1813-2001 disorder7 braoisg 11 1813-1978 grin
They all have their first appearance early in the corpus and continue to be in use
in the later corpus texts.
A further five lemmas have both <sg> and <sc> forms in the corpus but are clearly
shown in the results in the table below to be normally <sg>.
Table 7-7 Variation in <sc>/ <sg> formssc lemma text dates sg lemma texts dates gloss
1 feascair 1 1797 feasgar 142 1813-2007 afternoon2 truscan 4 1779-1828 trusgan 56 1813-2000 clothing3 ploscairtich 2 1804, 1979 plosgairtich 35 1828-2007 throb4 lascaire 2 1801, 1879 lasgaire 22 1801-1988 dandy5 ioscaid 2 1896, 1898 iosgaid 13 1841-1973 thigh6 pronnasc 6 1779-1973 pronnasg 12 1835-1979 sulphur7 brioscaid 7 1970-1989 briosgaid 13 1932-2007 biscuit
Again, the majority of <sc> forms are in the early corpus texts. Exceptions are the
relatively late uses of lascaire and ioscaid in the late nineteenth cenutry. The late
use of ploscairtich, 'throbbing', in Deireadh an Fhoghair (1979) is exceptional as it
is not used in an archaic context. For some other lemmas, including pronnasc and
brioscaid, the pattern is unclear with the corpus showing results over many
decades and in relatively small numbers.
149
The following table shows the lemmas which appear in fewer than 10 occurrences
in either form.
Table 7-8 <sc> and <sg> lemma with low frequencysc lemma no. of
textsdates sg
lemmano. oftexts
dates gloss
deasc 6 1970-90 deasg 3 1987-2005 desk
oscarra 3 1835-1968 osgarra 3 1978-2007 heroicmaicroscop39 2 1972-1976 - - microscopetoscair 1 1913 tosgair 4 1835-1972 ambassadordisc 1 1970 - - - diskbabhscadh 1 1896 - - - frightascart 1 1891 asgart 2 1923-1978 coarse lintguscul 1 1971 gusgul 6 1813-1988 refuseflasc 1 1971 flasg 7 1813-2005 flask
In these cases, the low frequency of the lemmas means that the norm cannot be
identified.
There remains a group of lemmas where <sc> appears to be more common which
are shown in Tables 7.9, 7.10 and 7.11.
Table 7-9 Dates for occurrences of bascaid and basgaidTexts 1800-1899 1900-1980 1981-90 1991-99 post-2000bascaid 3 12 3 2 0basgaid 0 3 1 3 3
Overall there are 20 occurrences of bascaid and 10 occurrences of basgaid,
however, the data does not indicate that the word is an exception to the norm.
Rather, it illustrates the continuing shift from <sc> to <sg>. During the nineteenth
century bascaid occurs with a <c> only. However, the last dated text to use it is
in 1994. It is basgaid with <g> that continues to be used after this into the twenty-
first century. Whereas the original loan follows a pattern of replacing English <k>
with Gaelic <c> (as in cidsin, 'kitchen'), the trend in the twentieth century has
been to replace that pattern with the phonological pattern of <sg> for /sk/,
demonstrating the regularising tendency of the last century that culminates in the
GOC recommendations of <sg> in all positions.
Some lemmas that resist the <sg> norm due to the prefix as- or eas- coming before
a lemma beginning with <c> (in the following Table 7.10). None of them have
39 This is listed as maicreasgop in the GOC 2005 & 2009 wordlists.
150
variants with <sg> appearing in the corpus; this would obscure the non-prefixed
forms of caraid, creidimh etc. which explains their resistance to norm of <sg>.
Table 7-10 Prefixed as- and eas- before <c>.Lemma No. of occurrences No. of texts Dates Glossascaoin 30 17 1779-2000 unkindeascaraid 32 18 1832-1973 foeascreidimh/eascreidimh
7 5 1835-1932 irreligious
eascruthach 1 1 1990 abstracteascòrd 1 1 1896 disagree
The corpus also contains forms of eas- followed by a hyphen. This would be the
preferred GOC-compliant form where the stress falls on the second syllable,
although this is not necessarily the case. Colin Mark's dictionary (2003) uses a
hyphen after eas-, but does not for ascaoin. The Co-fhaclair Gàidhlig (2011) and
the LearnGaelic.net dictionary list eascaraid as a headword.
There remains one more lemma that appears to be an exception to the <sg> norm
with no clear explanation.
Table 7-11 Frequency of deisciobalsc lemma No. of texts Dates sg lemma No. of texts Dates Glossdeisciobail 19 1779 -1995 deisgibal 1 1828 disciple
The word for 'disciple', deisciobail, only has one example of <sg> early in the
nineteenth century. Carswell used both in 1567, but the <sc> form has resisted
the widespread change to <sg> that other religious, and Latin or English loans such
as sgriobtair (scripture) have undergone.40 In MacLeod and Dewar (1839), there is
an entry for deisgiobul which redirects to the <sc> form deisciobul. Its only
appearance in the corpus with <sg> is notably aberrant as the text in which it
appears, Co'Chruinneachadh (1828) also uses deisciobul four times. It was not
listed in any of the GOC wordlists. While the recent corpus texts are not of a
religious nature where the word might be likely to be found, outwith the corpus
the <sc> form is still the norm in recent versions of the Bible41 and in 'Na Duilleagan
Gàidhlig', the Gaelic supplement of the Church of Scotland's Life and Work
magazine. It is also the form given in the Co-Fhaclair Gàidhlig (2011), in Mark's
40 Carswell uses descibul l.3314 and deisgibluibh l.1923 (line numbers in Thomson 1970).41 e.g. National Bible Society of Scotland (1992)
151
2003 dictionary and on the LearnGaelic.net dictionary. It is not clear why
deisciobal should resist the reform of <sg>: it could be that its religious
connotations and use in religious texts led to the spelling being conservative.
However, other lemmas with religious connotations, such as sgriobtair (scripture),
were fully adopted as <sg>.
7.3.4 Summary of <sc> and <sg>
It is clear then that <sg> in all positions has been the norm since the early
nineteenth century. Some conservative uses mean that <sc> still appeared
occasionally in the nineteenth century, particularly in the first half. However,
loans from English with k account for some later twentieth century occurrences
with only deisciobail strongly resisting the norm. The GOC recommendation for
<sg> in preference to <sc> therefore codified the conventional usage of modern
Scottish Gaelic.
7.4 The <sb> and <sp> clusters
7.4.1 Variability in <sb> and <sp>
The consonant clusters <sb> and <sp> both represent phonological /sp/. While
both were used in early Gaelic books, by the time of the earliest corpus texts,
<sp> was the more common form, the use of <sb> remaining only in a few words.
The variability in early Gaelic texts that was noted in reference to <sc> and <sg>
is also apparent in this case. In the Dictionary of the Irish Language (1984), there
are no headword entries beginning with <sb>; however, variation between <sb>
and <sp> is often attested in citations, e.g. in the entries for sprúileach,
'fragments', taisbénad, 'display', and cusbóir, 'subject'. According to R.L. Thomson,
initial <sb> was more common than <sp> in Foirm na n-Urrnuidheadh (1567)
(Thomson 1970: xiii). However, of the four lemmas which have initial <sb>,
sbēclāir, 'mirror' and sbor, 'spur', occur only once each and the other two show
variation: 'special' appears as sbesialta (3 times) and speisialta (5 times); 'spirit'
appears as sbiorad (27 times) and spiorad (13 times). As was the case with <sc>
and <sg>, the more frequently a word occurs, the more likely it is to appear in
both forms. Thomson does not give an overall picture of medial and final <sb> or
152
<sp> in the book but he does say that in easbul (apostle) they are both used 'about
equally' (Thomson 1970: xiii).
MacLeod and Dewar's 1839 dictionary redirects <sb> to <sp>, except for sbàirn
(struggle) and sbrogaill (crop, double chin) which would indicate that initial <sb>
forms were known yet atypical in the early nineteenth century.
A description of the use of <sb>/<sp> was not included in the 1981 GOC report. In
the 2005 report, <sp> was defined as the norm with a short list of exceptions using
<sb> explained by their 'frequency and familiarity':
The letters sp should be used in all positions in place of sb:_ cuspair, cuspann, speal, uspag
However, because of their frequency and familiarity, the spelling of thewords deasbad, easbaig, Gilleasbaig and taisbeanadh should be leftunchanged. (SQA 2005: 3)
The 2009 report amended the list of exceptions, removing Gilleasbaig, 'Archibald',
(which was retained in the wordlist) and adding susbaint, 'substance' (SQA 2009:
4). The attached wordlists in the 2005 and 2009 reports, however, also included
isbean, 'sausage' and leasbach, 'lesbian'. The 2009 wordlist also added cusbann,
'customs', as an alternative to cuspann.
We can expect the corpus to demonstrate that <sp> is the norm word initially. The
GOC recommendations suggest that variability in medial and final <sb> and <sp>
will be, at least to some extent, particular to different words.
7.4.2 Initial <sb> and <sp>
The corpus returned 59 lemmas beginning with <sp> from 6,337 occurrences and
only one lemma with one occurrence of initial <sb> in the corpus. The only <sb>
occurrence is sbruileach (fragments) which appears in 1804 in Eobhann
MacDiarmaid's Searmona [Sermons]. As was noted earlier, the DIL entry for
sprúileach has citations with both <sb> and <sp>. It is clear that in the period
covered by the corpus, 1750 to 2007, that initial <sp> is the norm.
153
7.4.3 Medial and final <sb> and <sp>
The corpus returned 26 lemmas with medial or final <sp> from 759 occurrences
and 16 lemmas with medial <sb> from 504 occurrences which indicates a mixed
picture with regard to norms here. The corpus returned no lemmas with final <sb>.
There are 16 lemmas which appear as <sp> only and they are listed in Table 7.12
below.
Table 7-12 Lemmas with only <sp> in the corpusLemma No. of
occurr.No. oftexts
Year Gloss
1 connspann 43 15 1813 - 1997 hero2 uspairn 10 9 1841 - 2004 struggle3 farspag 8 6 1841 - 1995 gull4 cusp 6 5 1841 - 1978 chillblain5 riaspach 5 5 1841 - 2000 disordered6 rusp 3 3 1813 - 2004 file7 coimhearspa 3 2 1841 - 1972 hanging back8 conspàirn 2 2 1841 - 1873 rivalry9 inspeactar 3 2 1970 - 1992 inspector10 ìosp 2 2 1813 - 1841 padlock11 correspondadh 1 1 2007 correspond12 duspainn 1 1 1973 uproar13 hanspeic 2 1 1944 (unidentified)14 prionspal 1 1 1932 principle15 taspallach 1 1 1923 witty16 ùspair 1 1 1911 ugly fellow
Examples 1 to 5 suggest that <sp> is the conventional norm for these lemmas.42
For the other lemmas which have low frequency in the corpus, there are some
loan words from English: prionspal, correspondadh, inspeactar, but too few
examples to establish any conventional pattern.
The following table, Table 7.13, shows all 16 medial <sb> lemmas and their <sp>
equivalents. For lemmas 1 to 5, it is clear that the <sp> form for these lemmas is
the norm. Unlike medial <sc>, however, there is not a clear chronological pattern
across the <sb> variants shifting from one form to the other. For lemmas 6 to 11,
the corpus returns were very low with only one or two occurrences; therefore no
conclusions can be drawn.
42 connspann, farspag, cuspa all appear in LearnGaelic.net.
154
Table 7-13 Lemmas with medial <sb> in the corpus and their <sp> variantssb lemma No. of
textsDates sp Lemma No. of
textsDates Gloss
1 cusbair 6 1779-1896
cuspair 96 1797-2007
subject
2 connsbaid 1 1828 connspaid 25 1779-2000
dispute
3 osbag 1 1990 ospag 17 1835-1995
sigh
4 osbadal 2 1970,1972
ospadal 16 1944-2007
hospital
5 cusbann 1 200143 cuspann 8 1813-1973
customs
6 presbiter 1 1898 - - - prebytery7 clisbeach 2 1938,
1970- - - unsteady
8 isbean 1 1987 - - - sausage9 basbair 1 1841 - - - fencer10 crùsbal 1 1896 - - - crucible11 connsbeach 1 1841 connspeach 1 1835 wasp12 taisbean 65 1779-
2007- - - display
13 deasbad 39 1804-2005
deaspad 1 1893 debate
14 easbaig 31 1779-1997
easpaig 5 1779-1898
bishop
15 susbaint 22 1813-2007
suspaint 2 1923,1932
substance
16 prosbaig 8 1891-2005
prospaig 1 1970 telescope
Lemmas 12-16 in the table have <sb> as the established norm, including the four
exceptions covered by GOC. Taisbean is clearly the established norm as no
spellings with <sp> appear at all. Deasbad has just one occurrence in a book of
poetry from 1893. Easbaig is slightly more mixed as the <sp> spelling does appear
in several nineteenth century texts, although even then it was not the dominant
form. Susbaint is both the older form, occurring in five nineteenth-century texts,
and the more common one.44 The corpus suggests prosbaig as another exception
to the <sp> norm. It does not appear in the GOC wordlists as an exception but
appears to be in widespread use. It is listed as prosbaig in Bòrd na Gàidhlig's Co-
fhaclair Gàidhlig (2011) and the LearnGaelic.net dictionary.
43 Used in Sgeulachdan Dhòmhnaill Alasdair [The Stories of Donald Alasdair] (2001). The Co-fhaclair Gàidhlig lists only cusbann as a headword although the 2009 GOC wordlist allows for both<sb> and <sp>.44 In this case, there is the possible influence of English substance that establishes <b>.
155
7.4.4 Summary of <sb> and <sp>
This cluster of <sb>/<sp> was the first of the three consonant cluster pairs to
standardise and settle on norms of usage as can be seen from the number of
occurrences. The relatively smaller number of lemmas with /sp/ may have been
a contributing factor to its regularisation. The use of <sb> word-initially ended at
the start of the nineteenth century. Medially and finally, although both forms are
used, they are lexically fixed with established conventions for using either one or
the other (cusbann/cuspann is the only debatable exception to this).
Comparing the corpus data to the GOC recommendations for <sb> to <sp>, the
recommendation and the list of exceptions in the reports are clearly following
existing conventions and their frequency and familiarity to readers. As the
exceptions where <sb> is the norm are few, GOC prioritises the conservative
spelling over simplification and uniformity that would break established practice.
7.5 The <st> and <sd> clusters
7.5.1 Variability in <st> and <sd> clusters
Both <st> and <sd> are used to represent phonological /st/. Variation in initial
<st>/<sd> was known before 1567; DIL has citations with <sd> forms for: stáb,
stábla, sdad, sdagún, staigre, stailc, sdair, stairide, stait, staitemail, stalcach,
stán, staraige, starga, stásion, stéd, etc. While in Foirm na n-Urrnuidheadh,
initial <sd> 'is regular, and st- occurs only in stiūradóir' (Thomson 1970: xiii), there
do not appear to be any <sd-> forms in Adtimchiol an Chreidimh c.1630 (Thomson
1962). Although medial <sd> is more common than <st> in Foirm na n-
Urrnuidheadh, Carswell is happy to interchange the two; if a lemma appears more
than 4 times, it is shown with at least one of <sd> and <st> (e.g. minisdir, freasdal,
criosd, feasda, baisd, ēisd).
The <st>/<sd> pair was the first consonant cluster to be explicitly revised by GOC.
It does so, according to the 1981 report, to address 'confusion' over the use of the
pair. We can assume this means that Gaelic school pupils were using the two forms
interchangeably or that writers were uncertain over the 'correct' form for a given
word. As there is no phonological difference between <sd> and <st> and as they
do not distinguish homophones, the 'confusion' is not likely to mean a lack of
156
comprehension, rather an inconsistency in usage which has become intolerable to
twentieth-century expectations. This 'confusion' over <st> and <sd> can be
illustrated with the example in the corpus of péist (reptile). It is used in one text
only, Bith-eòlas [Biology] (1976), but this text uses péisd in the singular (2 times)
and both péisdean (16 times) and péistean (3 times) in the plural, while listing in
its glossary péist - reptile.
The original GOC report recommended that:
There is some confusion apparent in the present practice of spellingwords with st/sd groups. Word-initially this is always <st>, and thisshould continue to be the case. Intervocalically and finally -st shouldbe used: for example astar, rithist. (SEB 1981: 2.2)
The revised recommendations in 2005, and repeated in 2009, added exceptions
and more examples to the <st>/<sd> pair:
The letters <st> should be used in all positions in place of sd:- aosta, a-rithist, èist, furasta, gasta, pòsta, staidhre, tuarastal,tubaist
An exception would arise in a compound place name where the finalelement is 'dal' or 'dail'. eg:
- Gramasdal, Lacasdal, Loch BaghasdailExceptions may also be found in the case of established forms ofpersonal names and nomenclature on signage, eg:
- Alasdair, Colaisde, Fionnlasdan, Taigh-òsda, Ùisdean- Likewise, Crìosd(a), Crìosdaidh etc.
(SQA 2005: 3A)
The compound place-name element where –dal or –dail follows an <s> is also
explicitly covered in the Ordnance Survey's Gaelic Names Liaision Committee.45
In the corpus, there is an example of confusion post-GOC where a strict adherence
to the recommendations can also lead to morphological anomalies. In Watson's
Saoghal Bana-mharaiche (2007), the singular sliasaid, undergoes syncope when
forming the plural, bringing the s and d together which the writer renders
sliastanan.
45 Replacing 'sd' with 'st' (SQA 2005 3A) does not apply at the junction of two distinct elements, e.g.Hìonnas + dal (from Norse dalr 'dale') = Hìonnasdal (Gaelic Names Liaison Committee 2006: 2)
157
Chunnaic mi bràthair dha is bha aon do na sliastanan aige air falbhdheth. Bha e air aon sliasaid. (Watson 2007: 32) (emphasis added)
I saw a brother of his and one of his legs was off. He was on one leg.46
7.5.2 Corpus Analysis of <st> and <sd>
In this section, a-steach and a-staigh have been considered as word initial <st>
lemmas. They are often written as a steach and a staigh in the corpus. They do
not have any <sd> forms in the corpus.
7.5.3 Initial <sd> and <st>
Carswell (1567) mainly uses initial <sd> even in loan words that have <st>, e.g.
sdaid, 'state', sdatūid, 'statute', sdiūradh, 'steer' and sduidēar, 'study'. The only
occasion where he uses <st>, in stiuradóir (steersman), occurs in a section that
appears to be an original composition by him (Thomson 1970: l.3879). However,
from then on, sdiùir not only seems to have been in common practice, it appears
to be the last lemma to switch to <st> in the mid-nineteenth century.
The use of initial <sd> is rare but does occur in the corpus in 2 lemmas. The
following table, Table 7.14 displays the occurrences from the corpus. There are
21 occurrences (no.s 4 to 6 below) from three published texts - MacMhaighstir
Alasdair's Ais-éiridh (1751), Leabhar Ceasnuighe Aithleasaighte (1779) and
Eobhann MacDiarmaid's Sermona (1804). Two more occurrences are references to
earlier manuscript forms cited in Eachann Bacach (1979).
46 sliastanan is glossed in the book as the dialectal word for casan, 'legs'.
158
Table 7-14 Corpus occurrences of initial <sd>Corpusoccurrence
Currentorthography
Year Location
1 sduidarrta stuidear c.1750 Eachann Bacach (1979)Appendix II, Moladh na Pìoba by IainMac Ailein (Ó Baoill 1979: 305).Form 'sduidarrta' appears in NLS MS72.2.15 'around 1750' (Ó Baoill 1979:xxxi)
2 sduidaradh stuidear mid-1700s
Eachann Bacach (1979)Appendix II, Moladh na Pìoba by IainMac Ailein (Ó Baoill 1979: 305)Form 'sduidaradh' appears in NLS MS72.2.2
3 sduir (x1) stiùir 1751 Ais-éiridh (Mac-Dhonuill 1751: 137)4 sdiuir (x 4) stiùir 1767 Tiomnadh Nuadh (1767)
5 sdiur (x 3) stiùir 1779 Leabhar Ceasnuighe Aithleasaighte(1779)
6 sdiùr (x 17) stiùir 1804 Searmona (1804)
Sdiuir is the only lemma MacMhaighstir Alasdair uses with <sd>, he uses 38 other
<st-> lemmas and it appears with <st> two other times as stiuir and stúradh.
Leabhar Ceasnuighe Aithleasaighte only uses <sd> for sdiuir. MacDiarmaid's
Searmona (1804) used <st> for sdiuir seventeen times. The following table shows
further examples of initial <sd> located outwith the corpus that extend into the
nineteenth century yet all consist of the same lemma, stiùir.
Table 7-15 Instance of sdiuir outwith the corpussd lemmas Year Locationsdiuirsdiuiradh
1778 (Shaw 1778: 8 & 10)An Analysis of the Gaelic Language
sdiùr 1816 (MacLaurin 1816: 13)Guide to the Reading of the Gaelic Language
fear-sdiuraidhsdiuradh
1857 ('Naigheachdan' 1857a: 8)An Teachdaire Gàidhealach (Australia)
Initial <sd>, then, survives the longest in sdiùir, the one lemma which Carswell
chose to use with initial <st> in stiūradóir (steersman).47 Over the first half of the
nineteenth century, this became the last lemma to change to initial <st>. An
illustration of this is made by Tormod MacLeòid, (Norman MacLeod) 'Caraid nan
47 The word is repeated seven times in lines 3877-93. R. L. Thomson believed this section to be an originalcontribution by Carswell (Thomson 1970: 170). While Carswell himself notes that the printer had noGaelic and so may have created errors (Thomson 1970: 112), it seems unlikely to account for theconsistent use of <st> in this section. Carswell used <sd-> consistently for 7 other lemmas: sdaid, sdàta,sdatūid, sdiūradh, sdoc, go sdrāsda, sduidēar.
159
Gàidheal', in an essay entitled 'Mu na Druidhean' which he included in two
collections. Where the first, Co'Chruinneachadh [...] ar son an Sgoilean [Collection
for Schools] (1828), uses <sd>, this is amended to <st> a few years later in Leabhar
nan Cnoc (1834):
[...] an uair nach robh riaghailt-sdiùraidh eil' aca' (MacLeòid 1828: 73-74) (bold added)
[...] an uair nach robh riaghailt-stiùraidh eil' aca (MacLeòid 1834: 235)(bold added)
It is likely that adherence to the Bible as a model for spelling led to this exception.
In Tiomnadh Nuadh (1767), there are 4 occurrences of sdiuir and none of stiuir in
the body of the Testament. In the 'Rules for Reading', however, stiuir is used twice
to illustrate vowels as 'stiuir, the helm', and 'stiuradh, steering' (Buchanan 1767:
7).
7.5.4 Medial and final <sd> and <st>
As it is clear that initial <st> established itself as the norm by the mid-eighteenth
century, it is in the medial and final position that the GOC recommendation of
1981 had its impact. The corpus returned over 200 lemmas with either <sd> or
<st>, most of which appear at least once in both forms. A large percentage of
these lemmas only appear in a small number of texts. Reliable results can only be
drawn from lemmas occurring at a reasonable frequency in the corpus; therefore
only lemmas appearing in more than ten texts are included in the following
analysis on intervocalic and final <st> and <sd>.
The corpus results show that the recommended change in use from <sd> to <st>
had significant impact. Given this, it is helpful to look at the results pre-GOC (i.e.
before 1981) and then to consider the results post-GOC (i.e. post-1981). Pre-GOC,
we can consider the extent to which norms can be seen. Post-GOC we can see how
and when the reform was implemented and whether any words have resisted the
change to <st> and any other anomalies.
160
The use of <st> and <sd> pre-GOC
This section analyses only pre-1981 texts. Only lemmas appearing in more than
ten texts are included in the analysis.
There are 21 lemmas which are predominantly spelt with <sd> which are shown in
the table below. One pattern which emerges is the preference for da in the
adjectival forms 15 to 21 in the table.
Table 7-16 Lemmas predominantly <sd>sd lemmas No. of texts st lemmas No. of texts Gloss
1 an-dràsda 120 an-dràsta 14 now2 èisd 152 èist 8 listen3 dleasdanas 34 dleastanas 0 duty4 am-feasd 36 am-feast 0 forever5 tuarasdal 37 tuarastal 0 wage6 taigh-òsda 47 taigh-òsta 0 hotel7 freasdal 63 freastal 0 attendance8 tosd 66 tost 1 silent9 feisd/feusd 30 feist/feust 2 feast10 asda 42 asta 2 out of them11 aisde48 55 aiste 15 out of her12 misde 47 miste 4 bad13 pàisde 51 pàiste 4 child14 tasdan 35 tastan 6 shilling15 gasda 85 gasta 6 excellent16 cneasda 40 cneasta 1 humane17 aosda 46 aosta 1 old18 blasda 43 blasta 2 tasty19 furasda 76 furasta 2 easy20 gleusda 57 gleusta 12 tuned21 pòsda 68 pòsta 3 married
<st> norms
There are 21 lemmas which are predominantly spelt with <st> (Table 7.17). There
are two general patterns that can be seen in these results. The first group are
those lemmas which have past participle -te endings. The second group are loan
words that, in their English or Latin forms are spelt with a <t>. There remain,
48 i.e. the prepositional pronoun. Aiste meaning 'essay' was counted separately.
161
however, lemmas such as tubaist, astar and fhathast that do not obviously
demonstrate any extrinsic reason for an <st> over <sd> spelling.
Table 7-17 Lemmas predominantly <st>sd lemmas No. of texts st lemmas No. of texts Gloss
1 laisde 1 laiste 32 lit up2 glaisde 0 glaiste 32 locked3 brisde 1 briste 52 broken4 brisd (verb) 7 brist (verb) 83 to break5 absdoil 0 abstoil 34 apostle6 baisd 1 baist 38 baptise7 maighsdir 1 maighstear 92 mister8 cisde 4 ciste 66 chest9 caisdeal 0 caisteal 69 castle10 minisdear 1 ministear 69 minister11 ceisd 36 ceist 87 question12 teisd 3 teist 43 testimony13 fhathasd 37 fhathast 128 still14 àbhaisd 2 àbhaist 109 ordinary15 a-rithisd 32 a-rithist 98 again16 asdar 8 astar 65 speed17 tubaisd 7 tubaist 39 accident18 - imcheist 24 uncertainty19 ròsd 15 ròst 24 roast21 goisdidh 1 goistidh 10 hair
Latin loan words
Early scribes of Gaelic were familiar with Latin and the spelling of borrowings may
have been influenced by the Latin spelling, establishing norms that would continue
even when the writers were no longer familiar with the Latin origin or spelling.
Table 7.18 below shows Latin origins as suggested by MacBain in his Etymological
Dictionary (1911) and from a list of Latin-derived loan words by Dr. George
Henderson in Dàin Iain Ghobha 2 (1896).
There are 13 lemmas of Latin origin of which 5 have no <sd> forms in the corpus:
abstoil, caisteal, imcheist, lòcust and mainistear. While this demonstrates that
the first three are clearly established norms, for lòcust and mainistear, however,
the results are not conclusive as there are only a small numbers of occurrences, 3
and 1 respectively.
162
Table 7-18 Loan words from Latin containing <st> or <sd>Lemma st texts sd texts Origin Ref.maighstir 116 1 magister MacBainceist 116 38 quœstio MacBainministear 94 1 minister MacBaincaisteal 90 0 castellum MacBainciste 84 4 cista MacBainbèist 70 35 bestia MacBainteist 53 2 testis MacBainbaist 44 1 baptizo MacBainabstoil 38 0 apostolus MacBainimcheist 36 0 from ceist MacBainlòcust 3 0 locusta Hendersonfeist 2 31 festia MacBainmainistear 1 0 monasterium MacBain
Four of the lemmas; baist, maighstir, ministear and teist only have <sd> forms in
1 or 2 texts which may be accounted for as simply minor inconsistencies:
baisd: one instance as aith-bhaisdeadh in Leabhar Ceasnuighe
Aithleasuighte from 1779 which is generally inconsistent in orthography.
The form baiste is used 6 times in this text as b(h)aisteadh.
maighsdir: one instance as maghisdar appears in a citation in the notes on
the edited text of the poem 'Upon the Revolution', taken from the
manuscript of Dr Hector MacLean (Dr Eachann MacGilleathain) MS -
MG15G/2/2 Public Archives, Halifax, Nova Scotia. The MS is believed to be
written in Mull between 1738-1768.
minisdear: appears 9 times all in the text of Dàin Iain Ghobha vol. 2 from
1896. The minisdear examples all appear in the same poem 'An Daoi agus
an Saoi'. This is unusual as ministear is used in the rest of the poems (it
appears 6 times). This irregularity may be caused by the fact that this book
was edited after the poet's death.
teisd: appears 3 times in 2 texts: Laoidhean agus Dain (1868) and Màiri Mhòr
nan Òran's Dàin agus Orain Ghàidhlig (1891). The latter text uses teist five
times and teisd once and the book has several printing errors which may
account for this irregularity. The former, however, only used teisd.
163
ciste: appears 4 times in 4 texts although two of these are different editions
of the same poem, 'A Mhàiri Ghaoil' by John Campbell of Ledaig. All the
texts with cisde also contain ciste.
There are two lemmas that, while still having a majority of <st> forms, also have
a significant proportion of <sd> forms, ceist and beist. Beist is, in fact, one of the
lemmas for which no norm is apparent from the corpus.
The only lemma in this group where the <sd> spelling was clearly the established
norm is feisd. The form feist only appears in An Teachdaire Ùr Gàidhealach (1835-
6) and K. C. Craig's edition of Sgialachdan Dhunnchaidh [Duncan's Stories] (1944)
which generally contains idiosyncratic and inconsistent spelling. This suggests that
writers were not influenced by the Latin spelling of festia.
No clear norm
The last lemma to be discussed here that appears in over 30 texts is béisd or biasd
which has no demonstrable preference for one form over the other. Nor is there
a clear distinction based on the spelling of the vowel as <èi> or <ia>. This
demonstrates that despite there being clear norms for a good number of the
lemmas, there are also words for which the picture is quite mixed.
Table 7-19 Bèisd and bèist in the corpussd lemma No. of texts st lemma No. of textsbéisd/biasd 33 béist/biast 54
The variation between these forms is evident in one of the earliest texts in the
corpus An Teachdaire Ùr Gàidhealach (1835-6) in which béisd, béist and biasd all
appear. As this text is a periodical with various authors' different spelling choices,
variation might be expected; however, it illustrates the options available and in
use at the time. One of the authors, 'C.C.', uses mainly béist or beist but on two
occasions uses béisd within the same story as the <sd> form. However, there are
only two stories that use biasd (none for biast) one by 'Tuathach Cuaileanach' and
one by 'Iain a' Chagainn', and both sign their locations as 'Bun Lochabar'.
164
English Loans
Similarly to Latin, many loan words from English that are spelled with <st> also
have <st> in their Gaelic equivalents. This does not apply universally, but it
accounts for many of the lemmas that contain <st> before 1981. English loan words
that have not been assimilated in any way into Gaelic spelling are not considered
here, e.g. capstan, investigations. I have considered them as Gaelic words if the
spelling has been altered significantly e.g. diostroidhear ('destroyer') or if Gaelic
morphology (plural endings etc.) has been applied.
In the following tables, Table 7.20 and 7.21, only pre-1998 texts are included. As
shown earlier, <st> can be expected in the corpus texts from 1998 where there is
largely adherence to the GOC recommendation. There are 21 loans from English
that appear as <st> without any <sd> examples in the corpus.
Table 7-20 English loans appearing as <st>Gaelic lemma English pre-GOC
texts1982-2007texts
1 astronautach astronaut 1 -2 chloroplast chloroplast 1 -3 ballaist ballast 6 -4 bairistear barrister 1 -5 blastadh blast 1 -6 ceimist chemist - 17 conastabal constable 2 28 cost coast 2 29 diosteampair distemper 1 -10 distriodhear destroyer - 111 faisisteach fascist - 112 firstaichean firsts 1 -13 forastair forester 1 -14 giustas justice 1 -15 hostail hostel - 116 ionnstramaid instrument 5 317 ostrais ostrich 1 -18 piostal pistol 7 -19 plastaig plastic - 120 siostam system - 221 tost toast 1 1
While some of these appear only in one text in the corpus, taken as a group they
demonstrate that the <st> cluster is usually retained in the Gaelic spelling.
165
There are also five English loans words with an <sd> form in the corpus before the
GOC recommendations as Table 7.21 shows.
Table 7-21 English loans appearing with both <sd> and <st>English Gaelic pre-GOC
texts1982-2007texts
breakfast bracoisd 1 0bracaist 8 6
canister49 canasdair 1 0canastair 8 2
crystal criosdal 1 0criostail 6 2
Protestant Pròsdanach 1 1Pròstanach 6 1
bolster bobhsdair 0 1bobhstair 3 1
For all of these lemmas derived from English loans, however, the <st> form isthe most frequent in the corpus.
Representing /ʤ/
There is a group of words where there is neither <st> or <sd> in the corresponding
English word but where <st/sd> has been used in the Gaelic spelling to represent
/ʤ/ in English words. Unfortunately they do not appear in significant enough
numbers to reveal a pattern. It seems that there has been no consistent approach
to rendering /ʤ/ in Gaelic.
49 According to the OED from Latin canistrum, from Greek kanastron in English from 15th century.However, unlike the religious words, it seems more likely that it came into Gaelic from English.
166
Table 7-22 English loans with /ʤ/Lemma pre-GOC
texts1982-2007texts
Gloss
colaisde 20 11 collegecolaiste 2 4sèisd 9 2 siegesèist - -bunndaisd - - poundage
(forbounty)bunndaist 2 0
mairisde - - marriagemairiste 3 1lòisdean 1 0 lodginglòistean 1 0curaisd 2 0 couragecuraist 1 0damaisde - - damagedamaiste 3 0seirsdean 1 0 sargeantseirstean - -
Post-GOC
As it is clear that the GOC reform of 1981 makes such a radical break with the
mixed picture of usage found previously, the change from <sd> norms to <st>
norms allows us to see when the GOC reforms were widely taken up. It can be
seen as a kind of shibboleth denoting whether a text has implemented the reform
or not. The following graph charts the number of types of <sd> and <st> lemmas
in each text in a given year in medial and final position:
Figure 7-1 Texts using <sd> and <st> 1980-2007
0
5
10
15
20
25
30
35
40
45
50
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
No. o
f tex
ts
Year
st
sd
167
The 1981 GOC report was not intended to have an immediate impact but to be
used in exams for the first time in 1988. The transition to GOC outside the school
system into wider publishing is suggested by this example to take place roughly
ten years after that.
An-dràsta
The change can be further detailed by looking at the data for an-dràsta, which
was predominantly spelt with <sd> before the reforms.
Table 7-23 Frequency of an-dràsda/an-dràsta in the corpusFreq. texts (total) texts pre-1981 1981-97 1998-2007
an-dràsda 687 139 120 19 0an-dràsta 164 22 13 0 9
It is important to note that this only demonstrates the picture in the corpus texts,
however, and simply indicates a general picture of real-world implementation of
the reform. In reality there would have been a more gradual cross-over period
between pre-GOC and post-GOC texts in the mid- to late-1990s.
There remain, however, some <sd> spellings in the corpus after the main shift to
<st> in 1998. Table 7.24 below gives the post-1998 texts that still include <sd>
forms.
Table 7-24 Lemmas containing <sd> post-1998Text st lemmas sd lemmas non-GOC compliantNa Klondykers (2005) 35 1 gun fhiosd' dhaTocasaid 'Ain Tuirc (2004) 46 2 casd (x 2)
tosd (x 2)Sgeulachdan DhòmhnaillAlasdair (2001)
37Incl.tuarastal x2
2 dleasdanastuarasdal
Bàrdachd DhòmhnaillAlasdair (1999): poems
18 2 dleasdanasfreasdal
Bàrdachd DhòmhnaillAlasdair (1999): intro
10 2 tasdancolaisde (used inproper noun)
Of the non-GOC compliant words, we can account for colaisde ('college') as an
exception covered by GOC to be acceptable in titles of colleges. While tasdan and
tuarasdal are rare, dleasdanas is relatively common and appears to resist the
reform. The instance of casd is a syncopated variant of casad. Instances of syncope
can create sd/st clusters and are worth considering in detail.
168
Syncope
There are five lemmas where the <sd>/<st> cluster has arisen from syncope (see
Table 7.25). There are not large numbers of these, but with the one exception of
sliasaid - sliastanan already mentioned, the same consonant used in non-
syncopated forms is retained. This accounts for some of the post-1998 uses of
<sd>: casd which occurs twice in Tocasaid 'Ain Tuirc (2004) and gluasd in Saoghal
Bana-mharaiche (2007).50
Table 7-25 <sd> and <st> clusters created by syncopeNon-syncopatedform
Lemma pre-GOCtexts
1982-1997texts
1998-2007texts
Gloss
casad casd 8 3 1 coughcast - - -
gluasad gluasd 3 1 1 movinggluast - - -
sliasaid slèisde51 14 0 0 thighslèiste 0 0 1
sluasaid sluaisdean 1 0 0 shovelsluaistean - - -
gaoisid gaoisd 1 0 0 hair ofbeasts
gaoist - - -
These examples illustrate competing principles in orthography: a strict
phonographic application of the 'one sound, one letter' rule that would require
only one combination of <st> and <sd> and the desire for the reader to able to
understand the connection between the non-syncopated and syncopated form.
Given that the corpus demonstrates that for some words the <sd> or <st> form
was an established norm, it might seem that there was an option for the reforms
to codify words as either <sd> or <st> forms. However, the corpus also contains
over one hundred lemmas that have no clear preference for one form over the
other. Attempting to choose which form a word should have would lead to an
extensive wordlist which was not GOC's aim. It would also mean that teachers
50 This is glossed in the text as 'Tha i gluaisd, .i. ‘tha i a’ gluasad’ ' (Watson 2007: 183).51 Both the genitive singular slèisde and the plural slèisdean have syncope and are counted together as the
same lemma here.
169
would need extensive wordlists to drill school pupils in which words should be
spelt with <sd> and which with <st>.
7.6 Summary and Conclusions
In this chapter, it has been established that variation between the pairs of
consonants clusters <st>/<sd>, <sc>/<sg> and <sp>/<sb> existed in the early texts
published in modern Scottish Gaelic. As printing in Scottish Gaelic significantly
increased with the evangelical publications of the later half of the eighteenth
century, variability in these pairs of consonants clusters was reduced, sometimes
to prefer one form only, sometimes lexically fixing the variation. The chronology
of change for each pair is different; however, they follow the same path where a
norm is established word-initially before the word medial and final positions are
settled.
The first of the pairs to standardise word initially was <sb> and <sp>, with <sp>
established as the standard convention before the period of the corpus. When
exactly the transition is made requires further investigation. For <st> and <sd>,
initial <st> is also the settled convention by the mid-eighteenth century. As there
are texts, both published and unpublished, in the first half of the eighteenth
century which are not yet included in the corpus, further work could establish
more precisely when this convention is introduced. The exception of sdiuir, 'steer',
is remarkable in this context for its non-conformity. Initial <sg> is also settled on
as the norm during the eighteenth century.
In medial and final <sc>/<sg>, the patterns of use show a clear preference for
<sg>. The pattern for non-conforming lemmas which use <sc> is that they are loan
words from English, particularly where the pattern of adapting <k> as <c> is
followed e.g. deasc 'desk', flasc 'flask', or bascaid 'basket'. The only other GOC-
resistant word is deisciobul.
In medial and final position, the use of <sb> and <sp>, which appear in a relatively
small number of lemmas, was again settled first and became lexically fixed. For
<st> and <sd>, the variability did not settle in all lemmas. The greater number of
lemmas compared with <sb>/<sp> made it more difficult for a pattern to be
170
lexically fixed. The first GOC report only covered <st> and <sd> as this was the
feature whose variability was causing difficulty.
Cox's criticism of GOC (2009) includes the different choices made with regards to
<sb>/<sp> and <sd>/<st>. Cox is not satisfied with the explanation in GOC 2005
and 2009 that frequency and familiarity allow for <sb> exceptions:
The claims of frequency and familiarity in the use of -sb- in these formsseem to be largely spurious, however, otherwise it is difficult to seewhy other words were not also exempted on similar grounds (e.g.aosda, now aosta). (Cox 2010: 163)
Cox considers 'frequency and familiarity' spurious because the same decisions over
<sb>/<sp> were not made in regards to <sd>/<st>. The exceptions to <sp> are
frequent, familiar, but also crucially of a small number. The data shows that the
much larger number of <st> and <sd> lemmas would be problematic in allowing
for regularised variation. There is a parallel with the difficulty of literate English
speakers with the difference between its and it's. It is not axiomatic that
guidelines or rules that require the average person to distinguish by grammatical
part of speech or phonological distinction are universally applied.
It is, of course, also relevant that the decisions were made at different times by
different people, in response to different circumstances.
What the form of these s+consonant clusters represents is the culmination of a
standardising process that demonstrates an ever-increasing attachment to
consistency, clarity and uniformity.
171
8 Simplicity vs Phonography: the accent
In this chapter I will trace the written description in scholarly and literary works
of how the grave and acute accents function in modern Scottish Gaelic, what
arguments are drawn on in discussion of their usage and then analyse the corpus
data to see how usage has matched or diverged from description.
The removal of the acute accent in the GOC reports was one of the most
controversial and discussed elements; however, discussion of its use has mainly
covered how it should be used; its actual use over the last three centuries has
never been documented. The primary ideological battle evidenced in the use of
the accent is between the phonographic ideal to represent distinct sounds, and
standardisation's drivers of simplicity, uniformity and consistency in usage.
As the primary interest is in the standardisation of the use of these two accents,
other accents used occasionally over the same period will not be included in this
study.52 Section 8.1 presents the written history and previous literature which
discusses the use of the accent. Section 8.2 covers the corpus analysis carried out.
Each vowel will be considered in turn, tracking its use and standardisation.
8.1 A History of the Accent in Scottish GaelicOrthography
The recommendation of the first Gaelic Orthographic Conventions (GOC) regarding
accents was one of the most contested points of those reforms. The 1981 report
advised the use of the grave accent only, foregoing the use of the acute. Notable
contemporary resistance to this recommendation includes the Gaelic columns of
The Scotsman newspaper, edited by Ronald Black and widely-used online
dictionary Am Faclair Beag, edited by Michel Bauer. The standard description of
the use of the accents prior to the GOC reform is as follows:
52 For example, the use of the breve on <ă> and the diaeresis or trema on <ë>, notably by JohnMacKenzie in Sar-Obair nam Bard Gaelach (1841) and Eachdraidh a' Prionnsa (1844), e.g.,Găël (Mac-Choinnich 1844: 122). Also in Laoidhean Spioradail (Grant 1862), Leabhar naFeinne (Campbell 1872), and Am Filidh Gaidhealach (MacKenzie 1873). Further examples canbe found by searching Corpas na Gàidhlig online.
172
1. Grave accent used on <a>, <e>, <i>, <o>, <u> to indicate a long vowel:
<a> /a/ <à> /a:/<e> /e/ <è> /ɛ:/<i> /i/ <ì> /i:/<o> /ɔ/ <ò> /ɔ:/<u> /u/ <ù> /u:/
2. Acute accent: used on <e> and <o> to indicate a high-mid long vowel:
<é> /e: /<ó> /o:/
Exceptions to this include no accent used to represent /i:/ when <i> is followed
by <nn> or <ll> in word-final position, e.g. /i:/ in binn, till. This is the system
often considered the 'traditional spelling'.53 However, in the written descriptions
of Gaelic orthography it takes some time for this 'traditional spelling' to be
expressed as such.
In this section I analyse how the use of the accent has been described over the
centuries. From the literature, I have identified 5 positions on the use of grave
and acute accents seen in Modern Scottish Gaelic texts. They are:
1. only the acute accent to indicate vowel length2. only the grave accent to indicate vowel length3. the grave on all vowels, also the acute on <é> for /eː/4. the grave on all vowels, also the acute on <é> and <ó> for /eː/ and /o:/
respectively (i.e. the 'traditional' model)5. no accents at all
While many of the elucidations of these models can be found in grammars,
dictionaries and textbooks, there are many comments also made in introductions
and editorial notes. Generally speaking, the history of the accent in Scottish
Gaelic traces a path from position 1 to position 4, with 5 appearing sporadically.
As will be shown, these are sometimes presented descriptively and sometimes
prescriptively.
8.1.1 Position 1: the acute accent only
This position inherits the conventions of Classical Gaelic although accents were
not always used, or used consistently, in Classical Gaelic writings. In Foirm na n-
53 e.g. ‘The traditional spelling used both accents’ (Bauer 2011: 376)
173
Urrnuidheadh (1567), which adapted Classical Gaelic to Roman type, Carswell
used the acute accent primarily although seven grave accents also appear
(Thomson 1970: xviii-xxi). As Scottish Gaelic entered the modern period, this
convention of acute usage continued. It was explicitly advocated in the Rev.
Robert Kirk's Scottish Gaelic wordlist published in Nicolson's Scottish Historical
Library in 1702:
Where a Vowel is accented, it is to be pronounced long; as a GreekOmega (Nicolson 1702: 335)
When Kirk had, in 1690, published an edition comprising O'Donnell's and Bedell's
Irish Old and New Testaments in Roman script he had also used the acute accent
throughout. In the rare printed texts in Scotland in the early eighteenth century,
such as the collection of psalms Sailm Dhaibhidh [The Psalms of David] (Synod of
Argyll 1715), the acute is predominant and forms the convention.
As described in Chapter 3, by the middle of the eighteenth century Classical Gaelic
orthography was at odds with the impulse to spread literacy and the word of God
in vernacular Scottish Gaelic. Alasdair Mac Mhaighstir Alasdair's spelling reforms
took a large step away from these Classical conventions. However, in both the
publication of his Gaelic-English vocabulary, Leabhar a Theasgasc Ainminnin: A
Galick and English Vocabulary (1741), and his poetry, Ais-éiridh na Sean Chánoin
Albannaich (1751), Mac Mhaighstir Alasdair maintained the Classical conventions
of using the acute accent, describing its use as follows:
And also all the vowels that have this accent (΄) above them are to bepronounced long, in whatever part of a word it may occur. (Mac-Dhonuill 1751: ix-x)
There is only one instance of a grave accent, <ò> in mheòbhair (meomhair,
memory) (Mac-Dhonuill 1751: 8) and its exceptional appearance implies it was
used in error. Assuming MacMhaighstir Alasdair saw his own work through the
press, the use of the acute would have been his decision.54 The 1741 vocabulary
was published by Robert Fleming in Edinburgh who was not a Gaelic speaker and
did not publish Gaelic works in general. The publisher of Ais-éiridh (1751) is
54 The manuscript in the National Library of Scotland, Adv. MS. 72.2.13, which contains some ofMacMhaighstir Alasdair’s poems and is believed to be in his own hand is written in cló Gaelachand uses the acute accent. An image of the MS can be seen in Dressler & Stiùbhart (2012).
174
unknown. It is the only text in the corpus to follow the practice of using the acute
accent only.
Position 1, then, covers the period up to mid-eighteenth century. It is not
advocated or used conventionally after the 1750s.
8.1.2 Position 2: the grave accent only
In Kirk's edition of Bedell's Bible published in 1690 there is no commentary on
spelling. However, when it was republished by John Orr in Glasgow in 1754, it
came with a postscript entitled A Leughoir (Dear Reader) with guidance on spelling
conventions. Black attributes this to Orr himself (2010: 238). This is the first
instance I have located where the use of the grave accent is explicitly
recommended. It names the grave, acute and circumflex accents but then only
describes the use of the grave accent.55 Indeed it appears that the grave has taken
the place of the acute throughout the 1754 publication, despite no other obvious
orthographical or editorial changes.
Three of John Orr's other publications, Sailm Dhaibhidh (Macfarlance 1753 and
1765) and a catechism, Leabhar-ceist na Mathair [The Mother's Catechism]
(Willison 1752), also use only the grave, which indicates that Orr's use of it is
deliberate. However, this could make Black's suggestion that Orr was the
surreptitious printer of Ais-éiridh (1751) less likely as that text uses the acute
throughout.56
In Section 2.2.2, the use of orthography as an identity marker was discussed. It is
conceivable that the switch from the acute to the grave was part of the process
of marking Scottish Gaelic as distinct from its Classical Gaelic tradition and its
Irish counterpart. By this stage, however, the use of the Roman font in printing as
opposed to cló Gaelach was already a significant distancing factor. While it
55 ‘Tha an Sineadh-garbh, na chomhartha air an chuid Sin don fhocal is còir a Shineadh, mar so,tròcair, far am bheill an Sioladh (trò,) fada, agus (cair,) goirid, agus cuiridh se eidir-dhealughadheidir focail, ann san aon dreach as fheugmhais, gidheadh, ag am bheill seadh air leath.' (Kirk1754: [372]) [The grave is a mark on that part of the word that should be lengthened, such as,tròcair, where the syllable (trò) is long and (cair) short, and it differentiates between words, inthe same form, yet which have separate meanings.]
56 Suggested in Black (2010: 236).
175
remains a possibility, I have not found any convincing evidence that the use of the
grave was motivated by this identity function.
Of the texts in the corpus, only one nineteenth-century text uses only the grave
accent, Marbh-rainn air daoine urramach (1887). However, of the publication's 15
pages, the grave appears only on the first five, so the absence of the acute could
be a printing limitation.57
After John Orr's break with existing convention, Position 2 can be seen again (if
somewhat indistinctly) in Robert MacFarlan's Alphabetical Vocabulary (1795; see
Section 8.1.3, p.166), but then not used conventionally until the late-twentieth
century and the reforms of the Gaelic Orthographic Conventions. Use of the grave
accent only to mark length is recommended by all three GOC reports and is the
current norm.
8.1.3 Position 3: the grave with the acute for <e>
The first translation of the complete New Testament in 1767 took a clear position
on the accent, favouring the grave but with a distinguishing role for the acute on
<e>. The 1767 edition came with a guide to the Rules for Reading which sets out
Position 3 for the first time. These recommend:
à for /a:/ ì for /i:/ ù for /u:/ è for /ɛ:/, the Greek Eta, or <ai> in praise é as the Latin <œ> ò for long /o:/
The full description for the use of the grave and acute for the two long sounds of
<e> is:
1. E sounds long, like the Greek Eta, or ai in praise, in è or sè he, rèduring the time of
2. E likewise often sounds long, as the Latin œ, or the first e in scene,as ré the moon, cé the earth, té a woman, Dé of God, an dé yesterday.That this sound of e may not be confounded with the preceeding, it will
57 It was published in Dingwall by the Ross-shire Journal office.
176
be proper to mark it with a different Accent, either when it stands as avowel, or enters into a diphthong. (1767: [2])
The Rules note that long <o> has two values but does not use an acute accent to
differentiate them, merely noting the exceptions of mòr, ò, bò, lòn, stòl and 'a
few others' which are not specified (1767: [2]).
Using the corpus to count the instances of each accent in Tiomnadh Nuadh (1767),
it is clear that both the translators and the printers were able to reach a good
level of consistency in the use of the grave as the data in the following table,
Table 8.1, shows.
Table 8-1 No. of accented vowels in Tiomnadh Nuadh (1767)Grave No. of occurrences Acute No. of occurrencesà 6,162 á 2 58
è 6,607 é 715ì 1,603 í 0ò 3,401 ó 0ù 1,184 ú 0
The Rules for Reading also imply that the accent is only required for
differentiation between short and long vowels when it is required to avoid
ambiguity, not for the marking of length in a consistent fashion:
All the vowels are sometimes long, sometimes short. When there is anydanger of mistake, the long are generally marked with Accents; as mìlea thousand, tìr a country, sàil the heel, bàs death: the two last wordsare thereby distinguished from sail a beam, bas the palm of the hand.(1767: [1])
By the next edition in 1783, however, this focus on differentiation is removed and
the accent is more 'generally' used for length:
All the vowels are sometimes long, sometimes short. When long theyare generally marked with accents. (SSPCK 1783: [x])
This indicates another thread in the standardisation of accents: early texts
describe them as useful for differentiation between words which would otherwise
be homographs, but this then became codified as regular long vowel
representation. Not all subsequent texts followed the New Testament's guidance.
58 These appear as ‘láimh’ and ‘grádh’. These appear to be printing errors as with the grave theyoccur 73 and 149 times respectively.
177
William Shaw's Analysis of the Gaelic Language (1778) does not cover the use of
accents at all in his orthographic description. His use of both grave and acute
accents in the body of the Analysis does not follow that laid out in the Tiomnadh
Nuadh or any other apparent consistent position. He (or the printer) also uses
accents in unusual places such as secondary syllables, e.g. snamhàm (1778: 101).
In contrast, Shaw (1780) does not use any accents at all, or make reference to
them, in his dictionary published two years later.
In Robert MacFarlan's 1795 Alphabetical Vocabulary with its 'directions for reading
and writing the Gailic', the recommendation to use the grave consistently to mark
vowel length is expressed, but this is a return to Position 2 espoused by Orr in
1754 — MacFarlan does not make a distinction between <è> and <é> and uses both
in his example. Despite using the same examples as in the Tiomnadh Nuadh (1767)
he does not follow those earlier Rules in differentiating between /ɛː/ and /eː/.
E sounds like ai in praise, in é, or sè, he, rè, during the time of, &c.(MacFarlan 1795: 10)
It was at the turn of the century that Position 3 fully established itself as the
convention. The Rev. Dr Alexander Stewart's influential Elements of Gaelic
Grammar was first published in 1801, the same year as the completed Old
Testament.59 In it, he explicitly states that his orthography follows the 'Rules' laid
out in the Tiomnadh Nuadh (1767) and describes the Position 3 convention.60
Subsequently, Armstrong's dictionary (1825: ii) and MacLeod and Dewar's
dictionary (1839: 437) continued this position.61 The Highland Society's Dictionary
(1828) is prefaced by Stewart's grammar and therefore also follows Position 3.
59 Stewart's Grammar appeared in 5 editions and further reprints in the following 100 years, as wellas forming the basis of Cameron Gillies' Elements of Gaelic Grammar (1902).
60 'In explaining the sounds of the letters I have availed myself of the very correct and acuteremarks on this subject, annexed to the Gaelic Version of the New Test. 1767' (Stewart 1801: 3)
61 Although MacLeod & Dewar do note that there are two long /o/ sounds: 'O, with the accent overit, sounds long, broad and open, like o in Lord. In many words it sounds differently from this,being like o in score. It might be useful to distinguish these sounds by a difference of accent, asis done in the case of accented E, but this nicety has not been practised in composition'(MacLeod & Dewar 1831: 437)
178
Variations from Position 3, however, continued to be suggested. An early book for
teaching is MacLaurin's The First Book for Children, in the Gaelic Language (1811).
In it, MacLaurin describes two accents to be used for two different purposes:
Tha da ghnè fhuaim fhada ann, fuaim ghéur fhada, agus fuaim chiùinfhada; tha comhar fa leth aca, tionndaidhear an comhar géur chum nalaimhe clithe, agus an comhar ciùin chum na laimhe deise, mar so ´ `ré, gnàth. (M'Laurin 1811: 6-7)
There are two kinds of long sounds, a long sharp sound, and a longsmooth sound; they each have a mark, the sharp mark is turned to theleft-hand side, and the calm mark to the right-hand side, like so ´ `ré, gnàth
However his use of the acute for 'fuaim ghéur fhada', the long sharp sound, with
<i> is unconventional:
Focail air an doigh chéudna [fòghair eadar da chomh-fhòghaire], far ambheil fuaim ghéur fhada aig an fhòghaire. tír, mír, cír, cís, bíd, dít, sír,sín, mín, rís. (M'Laurin 1811: 11)
Words in the same manner [a vowel between two consonants], wherethe vowel has a long sharp sound. Tír, mír (…)
The use of the acute on <i> was not continued by other writers.
Position 3, then, began with the Tiomnadh Nuadh (1767) and continued throughout
much of the nineteenth century. As the next section will shows, use of the acute
on <o> was slow to become established.
8.1.4 Position 4: the grave with acute on <e> and <o>
It was not until 1828 that James Munro's Gaelic Primer introduced the idea of using
the acute systematically on <o> using <ò> for /ɔ:/ ('Sounds like ô in for, lord',
Munro 1828: 5) and ó for /o:/ ('Sounds like ō in total, bold, foe', Munro 1828: 6):
Each of the five vowels admits of the grave accent `, which indicates along sound. The acute ´ is written over e and o only, and indicates aparticular long sound of these vowels. (Munro 1828: 2)
This arrangement of Position 4, which would then become the traditional position,
is laid out most clearly in a table of the alphabet where the list of 'powers' includes
179
the Gaelic and the corresponding English sound as the following: é = a in pate, è
= e in there, e = e in her, and ò = o in or, ó = cold, o = hot (Munro 1828: 30).
By the end of the nineteenth century this convention is repeatedly found in
descriptions of the orthography such as in MacBain's Etymological Dictionary
(1911: vii) and in Edward Dwelly's Gaelic-English Dictionary (1911: 376).62
8.1.5 Position 5: no accents
The earliest text in the corpus, Alexander MacFarlane's Gairm an De Mhoir (1750),
uses no accents. It was published by Robert and Andrew Foulis, Printers to the
University of Glasgow. Although when MacFarlane also published an edition of the
Synod of Argyll's psalms as Sailm Dhaibhidh (1753), it used the grave accent in
accordance with the practice of its publisher, John Orr.
Some writers have more explicitly preferred the removal of all accents. Alexander
MacBain hints at his dislike of accents in a discussion of standardisation in the
Celtic Magazine:
Uniformity in the spelling of Gaelic is much to be desired; indeed, it isan absolute necessity, if any good result is expected to come from theteaching of Gaelic in schools, to agree upon a standard spelling. [...]The apostrophes and accent must be dealt with firmly, and with a viewto their reduction, if their extinction is impossible, as perhaps it is.(MacBain 1887: 45)
In MacKinnon's essay on 'Accents, Apostrophes and Hyphens in Scottish Gaelic', his
first point about the accent raises the question of its removal altogether:
Should we not get rid of it altogether, as the English have done? Thiswould, no doubt, be the simplest solution. It would put an end at onceto all our difficulties and blunders in connection with the use of it. Butfew among us, I imagine, would vote for the total abolition of theAccent in writing Gaelic. If we had never had it, we would not miss it.But as things are it has been of advantage, especially to the youngreader (MacKinnon 1910: 195)
62 Other texts which describe Position 4: MacLennan (1925: xv) as a ‘rule’ of usage. Teachingmanuals; MacPharlain 1903, 1911 and 1913. MacFarlane calls the grave accent ‘iteag chlì’, leftplume, and the acute, ‘iteag dheas’, right plume (MacPharlain 1913: 3).
180
The most explicit argument for removing accents is made by Hector Cameron in
his edition of poetry from Tiree:
Throughout the book I have deliberately eliminated all accent marks.These, I agree, are entirely necessary for School Text-books, andcommendable for College Students' Class-books, but I do not think thatproficient readers of Gaelic really require them, especially in a book ofverse, where they have the assistance of rhyme and assonance. Whenwe are active walkers, we have no need to carry crutches. (Cameron1932: xxii)
Cameron agrees with MacKinnon in that accents are useful for the young reader.
Yet Cameron expects the general reader to have a fluency that makes the
additional phonological information provided by accents unnecessary. This
presumption of fluency is one of the most significant differences between
Cameron and more contemporary commentators on orthography who require
orthography to aid in phonological acquisition.
There are texts, including Cameron's Na Baird Thirisdeach (1932), without any
accents in the corpus although Cameron's is the only one where its absence is
explained and argued for.63 For most texts which contain no accents, printing
restrictions are as likely to be the reason as the deliberate choice of the writer.
8.1.6 GOC and the Removal of the Acute
When the first Gaelic Orthographic Conventions were published in 1981, they
returned to Position 2: grave only – not seen since John Orr's publications in the
mid-eighteenth century. The subsequent revisions in 2005 and 2009 repeated this
position. The clearest argument in print against the use of the acute is found in
Donald MacAulay's lecture to the Gaelic Society of Inverness, published prior to
the first GOC report. In it, MacAulay rejects the idea that distinguishing between
the two long sounds is necessary, presumably on the basis that a fluent speaker
uses the sounds regardless. Instead, he emphasises that greater simplicity and
uniformity would be possible with only one accent marking length:
it would seem a very good idea (especially in the light of the printingand publication problems we referred to above) to get rid of as many
63 Corpus texts with no accents are: Gairm an De Mhoir (1750), Dan Spioradail (1885) and NaBaird Thirisdeach (1932)
181
diacritic marks as possible. There seems no good reason to have twoaccent marks in Gaelic considering the very few word pairs they directlydistinguish. It would be sufficient to retain the grave accent simply toindicate length. (MacAulay 1976-78: 95)
The phonographic correspondence does not worry MacAulay – earlier in the same
paper he foregrounds the impossibility of the phonographic ideal and the inherent
differences between speech and writing. Instead MacAulay's focus is on the
practicalities of written communication. Thus, to support his argument that two
accents are not required, he cites the perceived ease with which readers can
approach texts without any accents:
That even this is not strictly necessary has been clearly demonstratedin the recent past by the omission of all accents from newspapercontributions in the Scotsman and in the West Highland Free Press, forexample, without any serious problems of comprehension arising.However, I would not (especially in view of the strong development ofGaelic as a second language) recommend the dropping of both accents,at least at this time. (MacAulay 1977: 95)
MacAulay's measure of success, according to this, is not how well the orthography
meets an ideal, but is in terms of 'comprehension', in terms of writing as a system
of graphic communication.
The complaints about the removal of the acute, however, have often been based
on the reduction of the phonographic correspondence. They are not, however,
simply aiming for an impossible phonographic ideal. Instead, there is a concern
for the effect on readers' phonological understanding of Gaelic, as in the example
below:
The traditional spelling used both accents because there is no otherway of predicting when you get either sound. Of course changing thespelling hasn't affected the pronunciation of these sounds! Well, fornative speakers anyway. Learners these days get this wrong quite a lotbecause few of the new books published use both accents. (Bauer 2011:376)
Opponents argue, albeit anecdotally, that there has since been a reduction in
quality of Gaelic pronunciation with Black noting 'a trend towards a single sound
for ò, è and even à,' due to the wider context where 'intergenerational
transmission ceased and orthography became crucial to the acquisition of the
language' (Black 2010: 254). For this reason, the acute is retained in the Fuaimean
182
na Gàidhlig: Introduction to the Sounds of Scottish Gaelic text used in teaching
at the University of Glasgow by Ó Maolalaigh 'in order to help learners identify
significant differences in pronunciation' (2010: 3) although they are advised not to
use it in writing and it is not used in other teaching materials. Similarly, Colin
Mark uses the acute to explain the graphic-phonic correspondence in his 2003
dictionary, but not throughout the rest of the dictionary. Current guidance on
function of the grave can be summarised by the following quote from a language
textbook:
The grave is mainly used to show that a vowel is long, but it alsodistinguishes vowel quality in some instances. (Watson 2012: 40)
The 'some instances' refers to the quality of <a> which is further detailed in 8.2.5.
8.1.7 The Practice of Accents
Despite the general consensus established in dictionaries and grammars around
the 'traditional' spelling, as represented by Position 4, there was also awareness
that its practice was not consistent. In 1910, MacKinnon had cause to complain
that contemporary practice 'depends mainly upon the caprice of the individual
writer' (MacKinnon 1910: 197). He also offers the explanation that the use of two
accents on <o> developed later as the vowels were not themselves consistent
across dialects:
Historically the close o has come to be marked by the Acute Accent onlyrecently, and by no means uniformly. The reason probably is that in thecase of e the two sounds are fairly consistent over the whole Gaelic-speaking area, while no such uniformity can be claimed for o.(MacKinnon 1910: 194)
In MacLennan's dictionary, we find the elaboration of Position 4 but with a caveat:
The application of this rule, however, is modified by the dialectalidiosyncrasy of any given writer. At the same time it may be taken asin the main of general value. (MacLennan 1925: xv)
According to the ideology of the standard, this inconsistency is a fault which
reflects 'caprice' and 'idiosyncrasy' and this is why writers prefer to simplify its
usage to better preserve the consistency that the ideology of the standard
requires.
183
The following corpus analysis looks at each of the five vowels in turn to see the
practice of accents in the chronological sweep that the corpus provides.
8.2 Corpus Analysis of Accent Use
Irish and non-Gaelic results were removed before analysis. The analysis focusses
on the most commonly used lemmas which are identified from the corpus data.
8.2.1 Accents with <e>
The corpus results for frequency of <è> and <é> can tell us in which lemmas the
accents are most common. By focussing on the most common lemmas, where most
data is found, general patterns of use can be deduced. Below are the five most
common lemmas across the corpus with their frequency. The corpus returned
20,722 instances of <è> and 41,494 instances of <é>. This is the only vowel where
there are more acute occurrences than grave. Of the <é> occurrences, 469 appear
in Ais-éiridh (1751).
The top five most frequent lemmas with <è> account for 42.1% of all <è>
occurrences. The top five most frequent lemmas with <é> account for 38.9% of all
<é> occurrences.
Table 8-2 Five most frequent lemmas with <è> grave<è>grave
freq. % oftotal <è>
<é> acuteform
freq. % oftotal <é>
dèan 3,928 18.9% déan 17 0.04%fèin 2,386 11.5% féin 7,190 17.3%dè 1,251 6% dé 2,053 4.9%an dèidh 959 4.6% an déidh 2,552 6.1%è 870 4.2% é 121 0.3%
Table 8-3 Five most frequent lemmas with <é> acute<é>acute
freq. % oftotal <é>
<è> graveform
freq. % oftotal <è>
féin 7,190 17.3% fèin 2,386 11.5%an déidh 2,552 6.1% an dèidh 959 4.6%éirich 2,322 5.6% èirich 525 2.5%céile 2,068 5.0% cèile 593 2.8%dé 2,053 4.9% dè 1,251 6%
184
Three lemmas, féin, 'self', dé, 'what' and an déidh, 'after', appear in the most
frequent list in both forms and are also more frequent in the corpus with the acute
form.64 The five most frequent <é> lemmas represent high mid vowels in most
dialects.
Féin and Fèin
The lemma féin is the most common one containing <é> for /eː/ and therefore its
use across the corpus can indicate the application of <é>. Table 8.4 show the
frequency of accented and unaccented fein before 1800.
Table 8-4 Frequency of fein before 1800Text féin fèin fein
Gairm an De Mhoir (1750) 0 0 430
Ais-éiridh (1751) 19 0 22
Tiomnadh Nuadh (1767) 0 0 1,171
Leabhar Ceasnuighe Aithleasuighte (1779) 0 0 56
Saighidear Criosduidh (1797) 0 0 40
However, in the eighteenth century corpus texts, fein is generally not marked long
at all. While Gairm an De Mhoir (1750) has no accents at all, the other four texts
make use of <é>. Even though Ais-éiridh (1751) only uses the acute, its use is
inconsistent and féin and fein both appear. While Tiomnadh Nuadh (1767) uses
<é> and describes its use in the Rules for Reading, it does not mark fein as long.
Leabhar Ceasnuigh Aithleasuighte (1779) follows this with unaccented fein.
Although Saighidear Criosduidh (1797) also appears to follow this, the text in fact
does not use any accents on <e> (although it does on other vowels).
The following table shows the occurrences in the nineteenth-century texts. The
beginning of the nineteenth century continues to show unaccented fein as the
most common form, with texts such as Daoine air an Comhairleachadh an aghaidh
bhi Deanamh Croin orra fhein [People Advised against Harming Themselves] (1832)
and Beachd-Chomhairlean [Hints for the Use of Highland Tenants and Cottagers]
(1838) consistently using no accent on fein. Searmona (1804) is the only text
where the grave fèin is in double figures before 1980.
64 Occurrences of an(-)dé, ‘yesterday’, and Dé, ‘God’, were counted separately.
185
Towards the middle of the century An Teachdaire Ùr Gàidhealach (1835-6) and
Sar-Obair nam Bard Gaelach (1841) are the first texts to use the acute féin in
significant numbers. This may follow the increase in the acute féin in dictionaries
including Armstrong (1825), Highland Society Dictionary (1828) and MacLeod and
Dewar (1831). The main variation after this is not between the grave and acute
but between the acute and the unaccented form, although MacEachen's 1842 and
MacAlpine's 1832 dictionary mark fèin with the grave.
Table 8-5 Nineteenth-century frequencies of feinText Féin Fèin FeinOrain Ghaelach (1801) 0 3 35Searmona (1804) 6 24 393Comhchruinneacha (1813) 10 0 155Co'chruinneachadh (1828) 4 7 285Daoine air an Comhairleachadh (1832) 0 0 195An Teachdaire Ùr Gàidhealach (1835-6) 356 5 16Beachd-Chomhairlean (1838) 0 0 121Sar-Obair nam Bard Gaelach (1841) 180 3 145Laoidhean Spioradail (1862) 42 1 22Laoidhean agus Dàin (1868) 42 0 1Am Filidh Gaidhealach (1873) 16 0 24Orain ann sa Ghailig (1875) 0 0 7Ordo Missæ (1877) 17 1 14An t-Oranaiche (1879) 195 2 13Gaelic Songs (1880) 11 1 0Poems and Songs (1880) 9 0 10Laithean Ceisde (1880) 51 4 2Poems (1884) 19 1 29Dan Spioradail (1885) 0 0 7Croft Cultivation (1885) 0 0 6Dàin agus Orain Ghàidhlig (1891) 60 0 11Dàin Iain Ghobha 1 (1893) 23 3 109An t-Urramach Iain Mac-Rath (1895) 0 0 98Dàin Iain Ghobha 2 (1896) 75 0 25Na Bàird Leathanach (1898) 9 0 82Leabhar na Ceilidh (1898): Bàrdachd 0 0 175Leabhar nan Gleann (1898) 14 0 1
The large number of texts where fein is unmarked by an accent may reflect
MacKinnon's observation that the 'rule' is influenced by dialects. However, texts
such as Dàin Iain Ghobha Vol. 2 (1896) and Poems (1884) are single author texts
which use both féin and fein.
As the next table shows, in the twentieth century the unmarking of the vowel in
fein significantly drops off with a handful of exceptions: Am Fear-Ciùil [The
186
Musicman] (1910), An t-Ogha Mor (1913), Na Baird Thirisdeach (1932) (which uses
no accents at all) and Litrichean Alasdair Mhoir [Letters from Alasdair] (1932).
The dictionaries of this time continued to codify the acute including MacBain (1911
[1896]), Dwelly (1911) and MacLennan (1925) and the use of the acute on féin
became more common. The very low numbers for the grave fèin suggests that
these are accidental typographical errors.
Table 8-6 Twentieth-century frequencies of fein pre-GOCText Féin Fèin FeinLaoidhean Bean Torra Dhamh (1906) 6 0 1Lòchran an Anma (1906): Urnaighean 75 0 0Oiteagan o'n Iar (1908) 183 1 0Am Fear-Ciùil (1910) 99 1 153Aig Tigh Na Beinne (1911) 257 3 1Dùn-Aluinn (1912) 349 0 0An t-Ogha Mor (1913) 13 0 163Spiritual Songs of Dugald Buchanan (1913) 50 0 0A' Bhraisd Lathurnach (1914) 57 0 1Cailin Sgiathanach (1923): sgeulachd 219 7 5Na Baird Thirisdeach (1932) 0 0 81Litrichean Alasdair Mhoir (1932) 0 0 600Highland Songs of the Forty-Five (1933) 38 0 1Gaelic Songs of Mary MacLeod (1934) 21 0 2Baird Chille Chomain (1936) 169 7 8Clarsach nam Beann (1937) 106 0 0Am Measg nam Bodach (1938) 196 0 0Songs of John MacCodrum (1938) 40 0 0Ban-altrumachd aig an Tigh (1939) 19 0 1Griasaiche Bhearnaraidh (1940) 9 0 0Sgialachdan Dhunnchaidh (1944) 242 4 1Folksongs and Folklore of South Uist (1955) 38 0 0Bùrn is Aran (1960) 73 1 0An Dubh is An Gorm (1963) 212 0 1Orain Iain Luim (1964) 31 0 0Sporan Dhòmhnaill (1968) 105 2 0Hebridean Folksongs 1 (1969) 20 0 0Dòmhnall Ruadh Chorùna (1969) 39 4 0Briseadh na Cloiche (1970) 110 2 6Luach na Saorsa (1970) 79 0 0An Clarsair Dall (1970) 21 0 0Tir an Aigh (1971) 289 0 4Lus-Chrun a Griomasaidh (1971) 174 1 1Saoghal an Treobhaiche (1972) 353 0 0Aitealan Dlù is Cian (1972) 122 0 0A' Bhratach Dhealrach (1972) 224 0 10Suathadh ri Iomadh Rubha (1973) 349 0 0Nach Neònach Sin (1973) 109 0 0An Aghaidh Choimheach (1973) 145 0 0Criomagan Ioma-dhathte (1973) 101 0 0
187
In the next table, the impact of the first Gaelic Orthographic Conventions of 1981
is very clear.65 There is a sudden shift the following year, beginning with Derick
Thomson's Creachadh na Clàrsaich (1982).
Table 8-7 Twentieth-century frequencies of fein post-GOC
In the post-GOC period, the adherence to the recommendation for the grave
accents is clearly reflected in the corpus data. Of the 27 texts that use at least
one form of fein between 1982 and 2007, only 6 use the acute accent and Fo Sgàil
a' Swastika is the only text to rely on the accentless fein. The use of the acute in
65 Here texts published in 1981 or after are grouped as 'post-GOC'. Including 1981 texts in thedata tables allows any contrast in results between the year of GOC's first publication andsubsequent years to be clearly apparent.
Creach Mhor nam Fiadh (1973) 77 0 0Fo Sgail a' Swastika (1974) 2 1 77Gàidhlig ann an Albainn (1976) 60 0 1An t-Aonaran (1976) 75 2 0Bith-eòlas (1976) 70 0 0Ugam agus Bhuam (1977) 211 0 0Hebridean Folksongs 2 (1977) 32 0 0Orain Dhonnchaidh Bhain (1978) 66 0 1Eachann Bacach (1979) 23 0 3Deireadh an Fhoghair (1979) 86 0 0Oighreachd is Gabhaltas (1980) 109 0 0
Text féin fèin feinHebridean Folksongs 3 (1981) 41 0 0Air Mo Chuairt (1982) 153 0 1Creachadh na Clàrsaich (1982) 0 35 0Suileabhan (1983) 0 96 0Seann Taighean Tirisdeach (1986) 0 19 0A' Ghaidhlig anns an Eilean Sgitheanach (1987) 0 0 3Air Druim an Eich Sgiathaich (1987) 3 211 0Sgriobhaidhean Choinnich MhicLeòid (1988) 0 416 0Dealbh-chruth nan Eilean Siar (1988) 0 0 3Spuirean na h-Iolaire (1989) 0 110 2A' Sireadh an Sgadain (1990) 0 146 1An Neamhnaid Luachmhor (1990) 0 130 0Bàrdachd na Roinn-Eòrpa (1990) 0 51 0Am Fear Meadhanach (1992) 0 164 5Coimhead air an Taigh-Dubh ann an Arnol (1994) 0 5 0Hiort (1995): Rosg 0 124 0Tuath is Tighearna (1995) 0 40 0Duanaire Colach (1997) 7 22 4Duanaire Colach (1997): intro 0 30 0Moch is Anmoch (1998): Donald A. MacNeill 5 0 0Bàrdachd Dhòmhnaill Alasdair (1999) 0 36 0Saoghal Bana-mharaiche (2007) 10 154 8
188
Moch is Anmoch (1998) is consistent with the texts' general non-conformity with
GOC. The appearances in Duanaire Colach (1997) are due to that text's use of
older quotes.
In the twenty-first century (Table 8.8), only in Smaointean fo Éiseabhal (2000),
both in Alex O' Henley's introduction to that book and in the poems themselves, is
féin used consistently. As with Moch is Anmoch (1998), Smaointean fo Éiseabhal
has other non-conforming spellings; which is unsurprising given that it was edited
by a critic of GOC, Ronald Black, who has consistently preferred the retention of
both accents.
Table 8-8 Frequency of fein in texts after 2000Text féin fèin feinSmuaintean fo Éiseabhal (2000): Am Bàrd 13 0 0Smuaintean fo Éiseabhal (2000): Roimh-ràdh 0 6 0Smuaintean fo Éiseabhal (2000) 8 0 0Sgeulachdan Dhòmhnaill Alasdair (2001) 0 102 0Dàin do Eimhir (2002) 0 8 0Tocasaid 'Ain Tuirc (2004) 0 190 0Na Klondykers (2005) 0 147 3
Using the example of féin, we can see standardisation being enacted through the
nineteenth and into the twentieth century. In this case, the Bible is not the
influential text. Instead, it appears to be the codification in the early nineteenth-
century dictionaries of the acute on féin. The twentieth-century results, despite
a small number of exceptions, shows a strong convention of acute é that is quite
consistent in the published texts. Its frequency was likely to mean that its spelling
was reinforced in the minds of readers.
8.2.2 Accents with <o>
There are 83,483 occurrences with <ò> and 12,330 occurrences with <ó>. Of the
<ó> occurrences, 510 appear in Ais-éiridh (1751). As was earlier noted, the grave
was the only accent recommended, and the only one used, for <o> in the
Tiomnadh Nuadh (1767). The suggestion to use both <ó> and <ò> is made in 1828.
The expectation would be for the corpus texts to use <ò> until roughly the 1830s,
without <ó> appearing regularly.
The five most common lemmas with the grave ò are shown in Table 8.9 along with
the comparative frequencies of the equivalent acute form.
189
Table 8-9 Five most frequent lemmas with <ò> grave<ò> grave freq. % of total <ó> acute
formfreq. % of total
mòr 4,285 5% mór 5,559 43.3%òg 3,834 4.4% óg 5 0.04%còir 2,967 3.4% cóir 11 0.08%beò 2,179 2.5% beó 8 0.06%dòigh 2,120 2.5% dóigh 2 0.01%
The five most frequent lemmas with grave <ò> account for only 17.8% of all the
<ò> lemmas in the corpus. The next table shows the five most frequent acute <ó>
lemmas and their equivalent grave forms.
Table 8-10 Five most frequent lemmas with <ó> acute<ó> acute freq. % of total <ò> grave freq. % of totalmór 5,559 43.3% mòr 4,285 5%vocableswith ó
1,648 13.4% vocableswith <ò>
1,042 1.2%
có 1,262 10.3% cò 808 0.9%móran 1,107 9% mòran 793 0.9%Mórag 413 3.3% Mòrag 104 0.1%
The top 5 acute lemmas account for 79.3% of all <ó> lemmas. The lemma mór
accounts for 43.3% of all <ó> lemmas. Overall, the acute <ó> is used in fewer
lemmas than <ò>.
An analysis of mór gives a general picture for the use of acute <ó>. The following
table shows the instances of mòr and mór in the corpus from the earliest text to
the end of the nineteenth century. After the expected norm of the acute in Ais-
éiridh (1751), it is the grave that forms the norm until the late nineteenth century.
190
Table 8-11 Frequency of mòr and mór from 1751 - 1900Text mòr mórAis-éiridh (1751) 0 53Tiomnadh Nuadh (1767) 447 0Leabhar Ceasnuighe Aithleasuighte (1779) 2 7Saighidear Criosduidh (1797) 1 0Orain Ghaelach (1801) 17 0Searmona (1804) 194 0Comhchruinneacha (1813) 18 0Co'chruinneachadh (1828) 266 0Daoine air an Comhairleachadh (1832) 5 0An Teachdaire Ùr Gàidhealach (1835-6) 113 314Sar-Obair nam Bard Gaelach (1841) 249 0Laoidhean Spioradail (1862) 60 0Laoidhean agus Dàin (1868): dàin 58 0Am Filidh Gaidhealach (1873) 22 0Ordo Missæ (1877) 4 1An t-Oranaiche (1879) 207 4Gaelic Songs (1880) 10 0Poems and Songs (1880) 31 13Laithean Ceisde (1880) 28 1Poems (1884) 11 1Dàin agus Orain Ghàidhlig (1891) 58 18Dàin Iain Ghobha 1 (1893) 58 1An t-Urramach Iain Mac-Rath (1895) 0 4
Dàin Iain Ghobha 2 (1896): Dàin 39 8Leabhar nan Gleann (1898) 22 83Na Bàird Leathanach (1898) 8 6Leabhar na Ceilidh (1898) 129 31
Leabhar Ceasnuighe Aithleasuighte's (1779) use of the acute generally is irregular;
the text contains seven grave <ò> in total and seven acute <ó> in total so its use
of the acute on mór here is not surprising. Dictionaries of the early nineteenth
century also use the grave mòr including Armstrong (1825), the Highland Society
Dictionary (1828), MacLeod & Dewar (1831), MacAlpine (1832) and MacEachen
(1842).
The use of the acute mór in An Teachdaire Ùr Gàidhealach (1835-6) is exceptional.
This corpus text covers 9 issues of the monthly periodical edited by Lachlan
MacLean (1798-1848) from Coll who was a shop-keeper in Glasgow (Thomson 1994:
181). It featured contributions from many parts of Gaeldom and this is reflected
in its inconsistencies in spelling and dialectal variants. As was noted in 8.2.1, it
was the first text to predominantly use the acute on féin. Section 8.2.5 will show
that it was the also first text to regularly mark the preposition á/ás, 'out of', with
the acute. After An Teachdaire Ùr Gàidhealach (1835-6), however, grave mòr is
191
still more frequent with acute mór only appearing in small numbers until Leabhar
nan Gleann (1898).
It is in the early twentieth century that acute mór becomes predominant (see
Table 8-12). This again aligns with dictionary recommendations: MacBain (1896,
1911) is the first to use the acute mór, whose usage Dwelly (1911) follows on this
occasion.66 Thereafter, dictionaries use the acute until GOC 1981. Two texts which
appear to resist the acute mór are Am Fear-Ciùil (1910) by Dòmhnall MacEacharn
[Donald Mackechnie] (1838-1908) from Jura, and Aig Tigh Na Beinne (1911) by
Katherine Whyte Grant (1845 -1928), a learner of Gaelic from Oban (MacBean
1921: 52-3). Grant says specifically in her introduction that she is not interested
in modernising her spelling:
Tha mi air cumail ris an t-sean dòigh-sgrìobhaidh. Tha clann chocleachdta ris, 'an leughadh a' Bhìobuill, agus leabhraichean eile, gu 'ndo mheas mi nach biodh e glic atharrachadh a dheanamh, no dòigheanúra a ghnàthachadh. (Grant 1911: 4)
I have kept to the old spelling. Children are so used to it, reading theBible, and other books, that I judged that it would not be wise to makechanges, or utilise new ways.
The relative age of these two writers might also therefore indicate a conservatism
in their spelling as they prefer the orthography they learnt when acquiring literacy
in Gaelic. Iain MacCormaic's use of acute mór in Oiteagan o'n Iar (1908) and Dùn-
Aluinn (1912) is notable for its consistency.67 Two of the other early novels, An t-
Ogha Mor (1913) and Cailin Sgiathanach (1923), also use the acute consistently.
The authors of these are Iain MacCormaic [John MacCormick] from Mull, Aonghas
MacDhonnachaidh from Skye, and Seumas MacLeòid [James MacLeod] from Harris.
MacKinnon's suggestion that the lack of uniformity in the pronunciation of <ò>/<ó>
compared to <è>/<é> is due to dialectal variation is of interest and would merit
further detailed investigation.
66 ‘mòr: see mór. Mòr is the spelling adopted by all Gaelic grammars and dictionaries exceptMacBain's and MacEachan's. MacBain's orthography has been adopted here.’ (Dwelly 1911:s.v. mòr)
67 Both were edited by Calum MacPhàrlain [Malcolm MacFarlane] from Lochaweside who used theacute mór in his School Gaelic Dictionary (1912).
192
Table 8-12 Frequency of mòr and mór from 1900 - 1980Text mòr mórLaoidhean Bean Torra Dhamh (1906) 7 0Lòchran an Anma (1906): Urnaighean 26 2Oiteagan o'n Iar (1908) 0 150Am Fear-Ciùil (1910) 54 5Aig Tigh Na Beinne (1911) 154 27Dùn-Aluinn (1912) 2 222An t-Ogha Mor (1913) 2 187Spiritual Songs of Dugald Buchanan (1913) 43 0A' Bhraisd Lathurnach (1914) 0 18Cailin Sgiathanach (1923): sgeulachd 2 128Litrichean Alasdair Mhoir (1932) 0 0Highland Songs of the Forty-Five (1933) 3 51Gaelic Songs of Mary MacLeod (1934) 0 24Bàird Chill-Chomain (1936) 32 30Clarsach nam Beann (1937) 30 20Songs of John MacCodrum (1938) 0 63Am Measg nam Bodach (1938) 9 166Ban-altrumachd aig an Tigh (1939) 2 37Griasaiche Bhearnaraidh (1940) 3 6Sgialachdan Dhunnchaidh (1944) 5 164Folksongs and Folklore of South Uist (1955) 0 46Bùrn is Aran (1960) 2 55An Dubh is An Gorm (1963) 0 85Orain Iain Luim (1964) 0 52Sporan Dhòmhnaill (1968) 1 102Hebridean Folksongs 1 (1969) 0 34Dòmhnall Ruadh Chorùna (1969) 3 18An Clarsair Dall (1970) 0 15Briseadh na Cloiche (1970) 0 111Luach na Saorsa (1970) 3 220Tir an Aigh (1971) 1 220Lus-Chrun a Griomasaidh (1971) 0 158Saoghal an Treobhaiche (1972) 0 132A' Bhratach Dhealrach (1972) 1 89Aitealan Dlù is Cian (1972) 0 195Suathadh ri Iomadh Rubha (1973) 3 491Criomagan Ioma-dhathte (1973) 6 42An Aghaidh Choimheach (1973) 0 102Nach Neònach Sin (1973) 0 63Creach Mhor nam Fiadh (1973) 0 43Fo Sgail a' Swastika (1974) 37 49An t-Aonaran (1976) 0 75Bith-eòlas (1976) 1 145Gàidhlig ann an Albainn (1976) 1 78Ugam agus Bhuam (1977) 0 118Hebridean Folksongs 2 (1977) 0 60Orain Dhonnchaidh Bhain (1978) 1 107Deireadh an Fhoghair (1979) 1 190Eachann Bacach (1979) 1 28Oighreachd is Gabhaltas (1980) 1 119
193
After the first GOC report, the results, as with <è>, show a clear switch and
subsequent adherence to the grave, beginning with Ruaraidh MacThòmais'
Creachadh na Clàrsaich (1982).
Table 8-13 Frequency of mòr and mór post-GOCText mòr mórHebridean Folksongs Vol. 3 (1981) 0 46Air Mo Chuairt (1982) 0 165Creachadh na Clàrsaich (1982) 49 0Suileabhan (1983) 188 2Seann Taighean Tirisdeach (1986) 40 0Air Druim an Eich Sgiathaich (1987) 123 0Sgrìobhaidhean Choinnich MhicLeòid (1988) 233 0Dealbh-chruth nan Eilean Siar (1988) 1 0Spuirean na h-Iolaire (1989) 101 0A' Sireadh an Sgadain (1990) 97 0An Neamhnaid Luachmhor (1990) 83 0Bàrdachd na Roinn-Eòrpa (1990) 54 0Iasgach (1991) 1 0Croitearachd (1991) 1 0Am Fear Meadhanach (1992) 121 0Taigh-Dubh ann an Arnol (1994) 4 0Hiort (1995) 283 0Tuath is Tighearna (1995) 49 0Duanaire Colach (1997) 112 13Moch is Anmoch (1998) 0 5Bàrdachd Dhòmhnaill Alasdair (1999) 54 0Smuaintean fo Éiseabhal (2000) 2 24Sgeulachdan Dhòmhnaill Alasdair (2001) 114 0Dàin do Eimhir (2002) 19 0Tocasaid 'Ain Tuirc (2004) 96 0Na Klondykers (2005) 183 0Saoghal Bana-mharaiche (2007) 158 15
The same GOC-resistant texts, Moch is Anmoch (1998) and Smuaintean fo
Éiseabhal (2000) are again the exceptions in their maintenance of the acute.
8.2.3 Accents with <i>
The 1767 Testament's usage and Rules for Reading apply only the grave accent to
<i>. Unlike <e> and <o>, there is no phonemic distinction between different long
sounds of [i] in Gaelic, the long vowel having only the quality of [i:], meaning that
only one accent is required to distinguish [i] from [i:]. As noted earlier, the only
text found which advocates the use of an acute with <i> in modern Scottish Gaelic
is MacLaurin's First Book for Children (1811).
194
The corpus returns 56,020 occurrences of <ì> and 545 results of <í>. The switch
from acute to grave is very clear from the corpus texts and indicates the New
Testament, Tiomnadh Nuadh (1767) as the key influencer. The predominance of
<ì> is clear from the overall numbers of accented <i> in the corpus.
Table 8-14 Five most frequent lemmas with grave <ì>: whole corpus results<ì> grave Freq. Glossnì 2,492 thingsìos 2,360 downtìr 2,320 landrìgh 2,305 kingfhìn 1,757 self
Table 8-15 Five most frequent lemmas with acute <í>: whole corpus results<í> acute Freq. Glossmín 19 soft,
smoothfíor 18 truerí(ogh) 18 kingtír 16 land(s)í 13 she/her
As most of the corpus occurrences of <í> can easily be accounted for, this will be
covered first in the following section before the results for <ì> are more closely
examined.
<í> with the Acute
The corpus returns 545 instances of <í>. Of these 72.6%, or 396 occurrences,
appear in MacMhaighstir Alasdair's Ais-éiridh (1751) which uses only the acute
accent. If these are taken out of the total, there are only 149 occurrences. This
changes which lemmas are the most common, shown in the following table.
Table 8-16 Five most frequent lemmas with <í> lemmas: without Ais-éiridh (1751)<í> acute Freq. Gloss(s)í 10 she/herní 6 thingdeídh 6 afterrí(ogh) 5 kingtír 5 land
Another 56 are archaic spellings which appear in extracts from texts from the
period pre-1751. All but one of these appear in Duanaire Colach (1997) and
Eachann Bacach (2007) and Highland Songs of the Forty-Five (1933) where there
195
are extensive quotes and citations from older poetry. In the former two, <í> is to
be expected as the norm for their period. In all eight occurrences in Highland
Songs of the Forty-Five (1933) they are citations of the original spelling from Ais-
éiridh (1751).
There is one instance in a biblical quotation in Aig Tigh na Beinne (1911).68 This is
an arcane usage or an error as Tiomnadh Nuadh does not use the acute <í> at all.
The rest of the poem repeats sìth with the grave <ì>.69
This leaves 93 instances where <í> is used in modern Scottish Gaelic orthography.
The following table shows these along with the number of <ì> lemmas in the same
texts. By the start of the 1800s its use appears to be accidental. There is no
indication of any exceptional lemma and no evidence that any particular word
resisted the adoption of the grave accent. Rather it appears that occurrences of
<í> post-1767 reflect the inconsistency and errors typical of Gaelic printing.
Examples marked with † are words which do not contain a long vowel where the
acute is even more likely to be a printing error.
Table 8-17 Comparative frequency of <ì> and <í>
68 “Gleidhidh tu esan ann an síth iomlan aig am bheil ’inntinn suidhichte ort, a chionn gu’n do chuire a dhòchas annad.”—Isa. xxvi. 3.(Grant 1911: 274)
69 Aig Tigh na Beinne (1911) has other erroneous printings of the acute e.g. uses <ú> once on úra,and <á> on tháinig, náirich, áite, and agháidh.
Text <ì> Gravefreq.
<í> Acutefreq.
<í>lemmas
Leabhar Ceasnuighe Aithleasuighte(1779)
9 9 (s)í x 9fíneamhuin, íomhaigh
Orain Ghaelach (1801) 237 1 mínComhchruinneacha (1813) 324 2 Albuínn†
íreCo'chruinneachadh (1828) 772 2 eilthíreach
bríghDaoine air an Comhairleachadh(1832)
12 1 tír
An Teachdaire Ùr Gàidhealach(1835-6)
2192 11 cínntedí-chuimhneéirídh†Eírionnfíon
fíreantachdínnseadhmílenísíol
Beachd-Chomhairlean (1838) 1 1 tastaín†Sar-Obair nam Bard Gaelach (1841) 2485 1 cíobairLaoidhean Spioradail (1862) 322 8 bhí
chíclímílltear
mínnírí(ogh)síth
196
From these results, the grave <ì> is consistently the norm for all these lemmas.
<ì> with the Grave
There are 56,020 occurrences of <ì> in the corpus. The five most common lemmas
containing <ì> in the corpus are as follows with the number of occurrences with
the acute given for comparison.
Am Filidh Gaidhealach (1873) 289 1 mín-mhal'Orain ann sa Ghailig (1875) 1 3 beoíl†
ceíle†slabhraídh†
An t-Oranaiche (1879) 1761 1 díreadhPoems (1884) 266 7 críth
-eíginíme
síorruidheachdstiuír†tí
Dàin Iain Ghobha 1 (1893) 339 3 aodhaír†sabaídibh†tríd
Leabhar nan Gleann (1898) 158 7 beaírn†deíghidh†gníomhínnte
nabuídhstarsnaích†steallaíbh†
Lòchran an Anma (1906) 512 2 aírgníomh
Am Fear-Ciùil (1910) 725 9 caílincaillíchéightíochfhuaír†frídh
mípaídhirríumrunnaích
Clarsach nam Beann (1973) 369 1 goírid†Am Measg nam Bodach (1938) 81 3 tír-mor
míleGriasaiche Bhearnaraidh (1940):Bàrdachd
75 4 císní
sgríobdíreach
Griasaiche Bhearnaraidh (1940):Roimh-ràdh
2 1 díreach
Lus-Chrun a Griomasaidh (1971) 406 1 mi fhínA' Bhratach Dhealrach (1972) 449 1 níFo Sgail a' Swastika (1974) 232 2 deídh†Nach Neònach Sin (1973) 423 2 a reír†
an deídh†Criomagan Ioma-dhathte (1973) 526 6 deídh†
feín†leínne
An t-Aonaran (1976) 309 1 litríchean†Hebridean Folksongs 3 (1981) 628 1 fhín
197
Table 8-18 Five most frequent <ì> lemmas<ì> grave Freq. <í> acute form Freq.nì 2,492 ní 8sìos 2,360 síos 7tìr 2,320 tír 16rìgh 2,305 rígh 18fhìn 1,757 fhín 2
As was established by looking at the acute <í> data, each lemma is by far more
established with the grave than with the acute.
8.2.4 Accents with <u>
In a survey of previous descriptions of orthographic 'rules for reading' or
pronunciation guides, beginning with the examples of 'ùr fresh, ùmhal humble' in
the first New Testament (1767: [3]), there is no mention of an accent on <u> other
than the grave (ù) to indicate length. The only exceptions are those that advocate
the removal of all accents, e.g. the Rev. Hector Cameron in 1932. As with <i>, the
corpus texts show the switch from acute to grave in the late eighteenth century
and indicate that the 1767 New Testament was the main influence of this.
Corpus Results for <ù> and <ú>
The corpus returns 56,520 occurrences of <ù> and 486 results of <ú>.
Table 8-19 Five most frequent <u> lemmas<ù> grave Freq. <ú> acute Freq.sùil 4,134 rún 7cùl 2,680 úr 7ùr 2,266 dlú 5dùthaich 2,080 rúdan 570
cùis 1,601 dúthaich 4
As with <í>, 74%, or 360, of the acute occurrences are from Ais-éiridh (1751).
There are also 46 occurrences which are quotes or reproductions of older Classical
texts where the acute is expected: 17 from citations from Ais-éiridh (1751) in
Highland Songs of the Forty-Five (1933); 17 times in 'Rainn' in Eachann Bacach
(1979) and 10 occurrences in Duanaire Colach (1997).
70 All rúd* lemmas occur in Sgialachdan Dhunnchaidh (1944) as 'rúdan ruadh'.
198
All of the remaining 80 instances can be accounted for as being most likely printing
errors given that none of the instances found of <ú> are used consistently by
lemma. The following table shows the occurrences per text. The frequencies of
grave <ù> lemmas in the same text are shown for comparison. After Ais-éiridh
(1751) the next corpus text to use <ú> is Leabhar Ceasnuighe Aithleasuighte
(1779), which uses <ú> more frequently that <ù>. After this, acute <ú> quickly
becomes rare.
199
Table 8-20 Comparative frequency of <ù> and <ú>Text <ù> freq. <ú> freq. <ú> lemmas
Leabhar Ceasnuighe Aithleasuighte (1779) 3 9búthdíúltrún
dhúibhfúigheadhúngadh
Orain Ghaelach (1801) 383 1 tlús
Comhchruinneacha (1813) 454 2múrúrrainn
Co'chruinneachadh (1828)1,204
3cúisstiúirúmhal
An Teachdaire Ùr Gàidhealach (1835-6) 1,989 6cúiscliúdlú
grúaimMúnichsiúil
Sar-Obair nam Bard Gaelach (1841) 3,039 1 súth-chraobhLaoidhean Spioradail (1862) 777 1 cúram
An t-Oranaiche (1879) 2,519 2dlúthmúr
Laithean Ceisde (1880) 104 1 stiúirPoems (1884) 208 1 tlús
Dain agus Orain Ghàidhlig (1891) 1,103 3crúndlú
dúthaich
Dàin Iain Ghobha 1 (1893) 446 3búrddúisg
úrnaigh
Leabhar na Ceilidh (1898) 632 5fúcadhtriúir (x2)
leúmtús
Am Fear-Ciùil (1910) 790 7
búrrallnúairsúilstiúir
Naishapúr (x2)cúirt
Aig Tigh Na Beinne (1911) 1,018 1 úr
An t-Ogha Mor (1913) 873 1 beús
Sgialachdan Dhunnchaidh (1944) 270 15
crúncúcúisdúilglún
lúblúireachrúdansmúidtriúir
Saoghal an Treobhaiche (1972) 559 2cúnntairdúnadh
Fo Sgail a' Swastika (1974) 284 2cúlleúd
Criomagan Ioma-Dhathte (1973) 538 14
cúramco-dhiúbhdlúdúthaichfeúmar
púnndsiúcarsúilúr
Sgialachdan Dhonnchaidh (1944) and Criomagan Ioma-Dathte [Many-coloured
Fragments] (1973) have the largest number of <ú> occurrences out of the corpus
200
texts, but given the much greater use of the grave in each of the relevant lemmas,
they can be assumed to be printing errors. These two texts were printed by
Alasdair Matheson & Co. in Glasgow and Techmac, Edinburgh, respectively.
Neither were regular publishers of Gaelic and perhaps this unfamiliarity led to
greater errors. Similarly, Fo Sgail a' Swastika's (1974) two instances of cúl and leúd
and the two instances in Saoghal an Treobhaiche (1972), are apparent printing
errors. There are no instances after the 1970s in the corpus — quotes from
Classical or Older Gaelic excepted — which I would suggest is due to better word
processing and printing quality.
8.2.5 Accents with <a>
The early orthographic descriptions in the late eighteenth and early nineteenth
century give long /a/ only one quality, /aː/, and one accent, <à>, to signify it.
However, the use of an acute accent on <a> has also been recommended for the
preposition á/ás, 'out, out of', in order to distinguish it from other meanings of
<a>, such as the possessive pronoun a, 'his, her', the other preposition a, 'to, into',
the relative pronoun a, 'that, which, who', the definite article and the vocative
particle. The pronunciation of <a> when it means (out, out of) is also often a
clear, sometime referred to as 'open', [a] rather than a reduced unstressed vowel,
although it is not long.
Despite it being argued that Gaelic has an accent as a vowel length marker and
despite GOC's further simplification to only have the grave accent, in this case
vowel quality still matters due to the polysemic nature of a in Modern Scottish
Gaelic.
The earliest reference I have so far located to this use of the acute on <a> is in
the orthographic notes to the 1826 quarto edition of the Tiomnadh Nuadh (New
Testament) although the recommendation is not accompanied by any
explanation.71 The preposition is listed without an accent in the dictionaries of
Armstrong (1825), MacEachen (1902), MacBain (1911) and MacDonald and Renton
(1979); MacLennan (1925) has both accentless and acute forms of a/as. It is in
Dwelly's dictionary (1911) that it appears as a headword with the acute accent, á
71 'A for “as” out of, marked with an acute accent, as “á teine” out of fire’ (SSPCK 1826: iii)
201
for 'out of,' despite not featuring in his introductory orthographic description.
Other writers who mention it include Alexander MacLean Sinclair who decides that
its function as a distinguishing mark, rather than a length mark, means it is not
really an 'accent'.
An accent shows that the vowel over which it is placed is long. The markover a in á, out of, is not an accent; it is merely a sign that serves todistinguish á, out of, from other words formed by the letter a. (MacLeanSinclair 1901: 4)
GOC 1981 recommended the grave only and made no exceptions for the acute,
marking 'à, out of' in its word list. Derick Thomson's New English-Gaelic Dictionary
was also published in 1981. It was the first published dictionary to follow the GOC
recommendations. Despite specifying in the introduction that the 'new system' of
GOC means that the dictionary uses 'only the grave accent as a mark of length',
the acute was maintained for both <á> and <à> (Thomson 1981: v):
Both à/às and á/ás survive in this work, for 'out, out of, ex-, dis-' etc.(Thomson 1981: v)
This use of á/ás continued in Gairm, the Gaelic quarterly, under the editorship of
Thomson until the end of its run in 2003. The Faclair na Pàrlamaid committee
disagreed with this convention but agreed that an accent was useful 'to denote
the particular quality' and used the grave:
We took the view that the term for “out of” should be written with thegrave accent: “à / às”, to denote the particular quality of the vowelsound. (McNeir 2001: 14)
The GOC conventions in 2005 and 2009 also referred to the preposition explicitly,
noting its 'open' quality. As with Faclair na Pàrlamaid, the grave is used to mark
this quality.
However, the accent should be written on à/às (out of) and on àswhenever the vowel is open (às bith, às dèidh, às mo chadal) todistinguish it from as when the vowel is not open (as fheàrr, as t-earrach, as t-samhradh etc). (SQA 2005: 5 & 2009: 6)
Others continue, despite GOC, to maintain the acute for <á/ás> including the
online dictionary Am Faclair Beag which also maintains the 'traditional' (Position
4) distinctions between acute and grave on <e> and <o>. As will be seen in the
202
corpus results, the use of the acute on á/ás, 'out, out of', is a long-standing
convention.
Corpus Results for <à> and <á>
The corpus returned 134,925 instances of <à> and 4,193 instances of <á>. In the
following table of frequencies, á/ás is counted when it appears as a simple
preposition; compound prepositions are counted as distinct items.
Table 8-21 Five most frequent lemmas with accented <a><à> grave Freq. <á> acute Freq.dà 4,913 á(s) 2,649àite 4,755 ám72 388fàg 4,538 ás déidh 74thàinig 3,946 ás-ùr 53àrd 3,678 ás aonais 25
It is clear that <à> is more common than <á> generally. The majority of <á>
instances, 60% of them, are of the most common lemma, á/ás. The occurrences
in Ais-éiridh (1751) account for 22%, or 582, of these occurrences. A further 91
occurrences are archaic spellings reproduced from the Classical Gaelic period in
Highland Songs of the Forty-Five (1933), Eachann Bacach (1979), Duanaire Colach
(1997), Saoghal Bana-mharaiche (2007), Bàrdachd Shilis na Ceapaich (1972),
Leabhar nan Gleann (1898) and Hebridean Folksongs 1 (1969). The following table
shows all the <á> lemmas which appear with more than 10 occurrences in the
corpus. The second most common lemma, ám, appears in only five texts before
the twentieth century and is most used in the twentieth century.
Table 8-22 Frequency of <á> lemmas with >10 occurrencesLemma No. of occurrencesá(s) 2,649ám 388ás déidh 74ás-ùr 53ás aonais 25ádhradh 23ás leth 15bás 21áite 16áithne 13
72 All with the meaning 'time'.
203
The Acute on <á/ás>
When the Ais-éiridh (1751) and Older Gaelic forms are set aside there remain
3,520 occurrences of which 77% are the preposition á/ás. While other instances of
<á> gradually disappear in the first quarter of the nineteenth century, á/ás for
'out of' remains common across the decades.
Table 8-23 Frequency of a/as in nineteenth-century texts
From the table, it is clear that in the oldest text, Searmona (1804), the grave à/às
is the convention; however, quickly after that, and around the time of the 1826
New Testament where <á> is recommended, the acute becomes the more typical
convention. Ronald Black suggests that it may have been MacKenzie's Sar-Obair
nam Bard Gaelach (1841) which launched, among other innovations, the use of
the acute on á/ás, 'out of' (Black 2010: 248). It is, however, clear that An
Teachdaire Ùr Gàidhealach (1835-36) takes precedence, using the grave <às> on
2 occasions and the acute <ás> on 47 occasions. For the second half of the
Text à/às á/ásOrain Ghaelach (1801) 0 0Searmona (1804) 0 0Comhchruinneacha (1813) 0 0Co'chruinneachadh (1828) 0 0Daoine air an Comhairleachadh (1832) 28 1An Teachdaire Ùr Gàidhealach (1835-6) 2 47Beachd-Chomhairlean (1838) 0 0Sar-Obair nam Bard Gaelach (1841) 3 74Laoidhean Spioradail (1862) 1 1Laoidhean agus Dàin (1868) 0 15Am Filidh Gaidhealach (1873) 1 7Orain ann sa Ghailig (1875) 2 1Ordo Missæ (1877) 0 6An t-Oranaiche (1879) 3 126Gaelic Songs (1880) 0 2Poems and Songs (1880) 0 5Laithean Ceisde (1880) 7 0Poems (1884) 0 0Dan Spioradail (1885) 0 0Croft Cultivation (1885) 0 0Marbh-rainn (1887) 0 0Dàin agus Orain Ghàidhlig (1891) 0 0Dàin Iain Ghobha 1 (1893) 7 63An t-Urramach Iain Mac-Rath (1895) 0 2Dàin Iain Ghobha 2 (1896) 3 43Na Bàird Leathanach (1898) 0 0Leabhar na Ceilidh (1898) 0 87Leabhar nan Gleann (1898) 14 41
204
nineteenth century the acute á/ás is the norm, meaning that the codification in
Dwelly (1911) is following the convention.
In the twentieth century before GOC seen in the following table, the results do
not challenge the convention of acute á/ás. There is only one clear exception
where grave à/às is used exclusively in Lòchran an Anma (1906). This is difficult
to account for as Lòchran an Anma does use <á> in three other instances: once on
ám, 'time' and twice on máthair, 'mother'.
Table 8-24 Frequency of a/as between 1900 and 1980Text à/às á/ásLaoidhean Bean Torra Dhamh (1906) 0 1Lòchran an Anma (1906): Urnaighean 18 0Oiteagan o'n Iar (1908) 0 21Am Fear-Ciùil (1910) 4 33Aig Tigh Na Beinne (1911) 0 192Dùn-Aluinn (1912) 0 42An t-Ogha Mor (1913) 0 2Spiritual Songs of Dugald Buchanan (1913) 0 5A' Bhraisd Lathurnach (1914) 0 1Cailin Sgiathanach (1923): sgeulachd 0 51Gaelic Songs of Mary MacLeod (1934) 0 11Baird Chille Chomain (1936) 0 19Clarsach nam Beann (1937) 0 5Am Measg nam Bodach (1938) 0 95Ban-altrumachd aig an Tigh (1939) 0 2Griasaiche Bhearnaraidh (1940) 0 4Bùrn is Aran (1960) 1 28An Dubh is An Gorm (1963) 0 111Sporan Dhòmhnaill (1968) 0 74Dòmhnall Ruadh Chorùna (1969) 1 3Briseadh na Cloiche (1970) 1 79Luach na Saorsa (1970) 0 72Tir an Aigh (1971) 0 2Lus-Chrun a Griomasaidh (1971) 1 78Aitealan Dlù is Cian (1972) 0 140A' Bhratach Dhealrach (1972) 2 61Suathadh ri Iomadh Rubha (1973) 0 314Nach Neònach Sin (1973) 0 93An Aghaidh Choimheach (1973) 0 147Criomagan Ioma-dhathte (1973) 1 17Creach Mhor nam Fiadh (1973) 0 19Gàidhlig ann an Albainn (1976) 0 37An t-Aonaran (1976) 1 104Bith-eòlas (1976) 0 132Deireadh an Fhoghair (1979) 0 63Oighreachd is Gabhaltas (1980) 0 94
205
Whereas, with the other vowels, the adherence to the GOC recommendation was
clear from MacThòmais' Creachadh na Clàrsaich (1982), MacThòmais in this case
preserves the acute on á/ás as he did in his 1981 dictionary and in his editorship
of Gairm (see Thomson 1981: v).
Table 8-25 Frequency of a/as post-GOCText à/às á/ásHebridean Folksongs 3 (1981) 0 0Air Mo Chuairt (1982) 0 0Creachadh na Clàrsaich (1982) 1 114Suileabhan (1983) 236 0Seann Taighean Tirisdeach (1986) 7 0A' Ghaidhlig anns an Eilean Sgitheanach (1987) 0 0Air Druim an Eich Sgiathaich (1987) 0 0Sgriobhaidhean Choinnich MhicLeòid (1988) 192 0Dealbh-chruth nan Eilean Siar (1988) 0 0Spuirean na h-Iolaire (1989) 100 0A' Sireadh an Sgadain (1990) 61 0An Neamhnaid Luachmhor (1990) 127 0Bàrdachd na Roinn-Eòrpa (1990) 52 0Am Fear Meadhanach (1992) 153 0Coimhead air an Taigh-Dubh ann an Arnol (1994) 6 0Hiort (1995): Rosg 265 0Tuath is Tighearna (1995) 14 0Duanaire Colach (1997) 21 0Moch is Anmoch (1998): Donald A. MacNeill 0 2Bàrdachd Dhòmhnaill Alasdair (1999) 35 0Smuaintean fo Éiseabhal (2000) 1 16Sgeulachdan Dhòmhnaill Alasdair (2001) 84 0Tocasaid 'Ain Tuirc (2004) 171 0Na Klondykers (2005) 168 0Saoghal Bana-mharaiche (2007) 222 0
The switch after the initial GOC report which applied the grave across the board
is visible in the corpus results. The acute on á/ás still appears in post-GOC
publications albeit only in Creachadh na Clàrsaich (1982) by Derick Thomson who,
as previously noted, advocated keeping the acute in this case; and in Smuaintean
fo Éiseabhal (2000) edited by Ronald Black who has advocated maintaining the
acute in this case and on <é> and <ó>.
8.3 Summary and Conclusion
This analysis has shown how the codification and standardised description of
accents in Scottish Gaelic orthography developed from the mid-eighteenth
century. Firstly, John Orr's editorial notes attached to his 1754 reprint of Kirk's
206
Irish Bible introduced the grave as a vowel length marker on all vowels. The use
of the acute on <é> for /eː/ is codified in the 1767 New Testament's Rules for
Reading by the Rev. James Stuart (1767: [2]). The use of the acute on <ó> is
recommended in 1828 by James Munro in his Gaelic Primer. The use of the acute
on <á/ás> 'out of', is introduced in the 1826 Quarto Bible of the SSPCK, the only
codified use of an accent for vowel quality, not length. The settled description of
this codification continued from the late nineteenth century to the 1981 GOC
report which removed the acute accent and used the grave on all instances.
It has been shown in the corpus analysis that usage has not strictly adhered to the
codified, prescriptive accounts. However, it can be seen that the convention of
the use of the grave for long <ì> and <ù> was established directly after first Orr's,
then the Tiomnadh Nuadh's example, with only occasional printing errors resulting
in the acute on these letters after Ais-éiridh (1751) (or Leabhar Ceasnuighe
Aithleasuighte (1779) in the case of <ú>).
In the eighteenth and early nineteenth century, the codified descriptions of the
use of the accent come with little explanation or commentary to permit
ideological analysis. However, the main use of the accent as a phonological marker
is indicative of the movement towards graphic-phonic correspondence. This
clearly takes its cue from the vernacularisation principle that informed much of
the orthographical development in the same period. It is not surprising then, that
the main engine of vernacularisation, evangelicalism by the Presbyterian Church,
provides the texts which offer this codification. The graphic-phonic
correspondence also lies behind the shift in emphasis from the initial description
of the accent being useful 'when there is any danger of mistake' to its being
recommended for consistent use. As the editions of the Testaments revised the
translation, the orthography was further refined, with finally both acute and grave
accents allowing for more vowel representation to signal different qualities.
Through the twentieth century, however, the ability to represent phonologically
distinct sounds became less important than the drive towards standardisation and
its requirements for simplicity, uniformity and consistency in usage. This
culminates in the recommendation of GOC (1981) for the use of only one accent.
With the simplicity of only one accent, errors made by printers and writers are
reduced.
207
9 Conclusions
9.1 Main Arguments
Le paradoxe orthographie [...] est tant vaine & incertaine, que leprocès en est encore pendant, les uns suyvans la raison, les autresl'usage, les autres l'abus, autres leur opinion & volonté. Et toutefois nonconstans & de mesme teneur, mais dissemblables entre eux, voire à euxmesmes. (Barthélemy Aneau, Quintil Horacian (1551) quoted inCerquiglini 2004: 31)
The orthographic paradox [...] is so vain and uncertain, that theprocess is still unresolved, some following reason, others usage, othersmisuse, others their own opinion and will. And all the whileinconsistent but of the same content, dissimilar to each other, andeven dissimilar to themselves.
Cerquiglini says that Aneau, writing in 1551, points out 'la contradiction où se
trouve la science de l'orthographe', the contradiction where the science of
orthography is found (2004: 32). That orthography is not a straightforward
application of rules reflects Coulmas' observation that real writing is compromise,
historic and pragmatic (see section 2.1.4).
The aims of this study were to begin the work of charting standardisation in
Scottish Gaelic orthography within its sociolinguistic context. The thesis took the
view that the study of an orthography did not mean an assessment of what the
'best' spelling should be, but rather followed Coulmas' recommendation to study
writing as it is, not as it should be (Coulmas 2003: 16). Previously, analysis of
Gaelic orthography has generally been made when writers and scholars wished to
propose or recommend reform. Specifically, the study investigated two
complementary questions: Which language ideologies lie behind the choices to
argue for reform or continuity in the spelling of Scottish Gaelic? What can a digital
corpus tell us about how Gaelic has been spelled historically? The thesis thus
examined the history of modern Scottish Gaelic orthography using sociolinguistic
ideological analysis and a carefully designed corpus of texts.
208
9.2 Main Findings
9.2.1 Which language ideologies lie behind the choices to arguefor reform or continuity in the spelling of Scottish Gaelic?
It has been argued in this thesis that Gaelic is not unusual in having an orthography
that is the result of a battle of ideas. Any standard is created and maintained
within various ideological pressures that shape it by pulling in different directions
at once. As with most other writing systems, Scottish Gaelic orthography is a
mixed system of phonic-graphic correspondence along with aspects of
conservatism and the need for uniformity that standardisation requires.
Scottish Gaelic writers and scholars have valued the phonographic ideal and used
it to defend the language from outside attack. They have also understood the
limitations of the phonographic ideal and have sought to balance graphic-phonic
correspondence with the needs of: avoiding homographs; preserving grammatical
relationships; conservatism; and acceptable dialectal variation. Etymology, in the
sense of the oldest form of words, is not as valued as conservatism, in the sense
of maintaining the status quo familiar to readers.
It has been argued that a standardised writing system has particular importance
for Scottish Gaelic as a minority language as it legitimises its existence. Criticisms
of Gaelic spelling have used apparent inconsistencies to accuse it of lacking a
standard, which in European literate culture is a way of delegitimising the
language. Orthographic reforms in the mid-eighteenth century were also a way of
establishing autonomy from Irish Gaelic and legitimising Scottish Gaelic as its own
language. It has also been argued that the idea of the standard in Scottish Gaelic
is not as a socially elite 'correct' variety but rather that there is a popular notion
of a standard as a supradialectal common written form which facilitates
communication between equally valid dialects. The thesis proposes that writers
and scholars have seen dialectal spelling as acceptable variation within what has
been termed MacKinnon's paradigm: the contexts of poetry, song and dialect
material.
209
9.2.2 What can a digital corpus tell us about how Gaelic has beenspelled historically?
This thesis has demonstrated how a corpus linguistic methodology can be used for
the study of Gaelic orthography. In particular, it has shown how corpus data can
reveal conventions, established exceptions to those conventions and changing
trends — all of which challenge existing perceptions of Gaelic spelling 'rules'. The
now-available electronic corpus of the Digital Archive of Scottish Gaelic and
Corpas na Gàidhlig can be exploited as a source of empirical data and evidence
for Scottish Gaelic, as well as for its planned use as a resource for the Historical
Dictionary.73
In the data showing <eu>/<ia> variation, the corpus supports the central and
peripheral dialectal distinction as <ia> occurrences only appear in central (or
unknown) dialect texts and are largely limited to MacKinnon's paradigm of
dialectal material or poetry.
With the s + consonant clusters, the corpus data showed the patterns of use and
the conventions as they became established. It also revealed how exceptions can
persist, notably with stiùir, prosbaig, deisciobul and <sc> in loan words from
English. As an example of the identification of exceptions, this study has identified
lexical items which have proved resistant to the GOC recommendations:
exceptions to the <sg> for /sk/ pattern were found in deisciobail and following
the prefix eas- (eascaraid, eascòrd) and the codification of <sp> over <sb> was
also resisted in prosbaig.
The switch from the convention of the acute in Ais-éiridh (1751) to the convention
of the grave by the end of the eighteenth century can be seen in the corpus and
the impact of the 1981 GOC recommendation to remove the acute is seen almost
immediately in the corpus texts. The corpus revealed the extent to which the
acute accent on <á> has been used. All three of the case studies in this thesis give
evidence of a tendency for spelling to become less varied over time: <eu> in place
of <eu>/<ia>, only one grapheme for each s + consonant cluster, one accent in
place of two. Overall, the patterns showed a standardising process that
73 See www.dasg.ac.uk
210
demonstrates an increasing attachment to consistency, clarity and uniformity.
With all of the case studies the corpus shows general uptake of the
recommendations of the GOC 1981 report.
A corpus analysis can furnish evidence on which forms have previously been used,
what patterns are already at work in the conventions, and what exceptions are
understood, to allow the community to make informed decisions.
9.3 Further Research
In covering a period of almost three hundred years, the thesis has necessarily
centred on the broad patterns of orthographic standardisation. There are,
therefore, many areas where further research would be illuminating. There is
much more detail in the individual language commentators, translators and
grammarians to be documented, following up on the work of Hogg (2011) on Rev.
Dr Alexander Irvine and Roberts (2006) on Ewen MacEachen, particularly to
elucidate the approaches and ideologies influential in the critical period between
1750 – 1850. This study has considered commentaries and expressed ideals that
have been located in in a wide range of published material; however, the National
Library of Scotland and other archives are likely to contain a greater wealth of
information and would merit more fine-grained analysis in future research.
This research selected three particular features of Gaelic orthography for analysis;
the variation between <eu> and <ia> for (broken) long /e/; the variation between
s + stop consonant clusters and the use of the accent. However, there are many
other orthographic features which would be able to cast light on the picture of
Gaelic orthography over the centuries. These include the application of the 'broad
to broad, slender to slender' spelling rule and the representation of dialectal
features. MacKinnon's suggestion that the lack of uniformity in the pronunciation
of <ò>/<ó> compared to <è>/<é> is due to dialectal variation in the pronunciation
of long /o/ is also of interest and would merit further detailed investigation. An
expansion of the work carried out here on accented characters, to take in the use
of accents on secondary syllables, particularly in the adaption of loan words to
Gaelic orthography would also be valuable.
211
Further sociolinguistic study of the members of the speech community today could
reveal whether current preferences are the same as in the late nineteenth century
when dialectal indexing was acceptable in dialogue (in literature or drama) but
not in formal writing. The dominant ideology in the literature is that variation in
spelling is bad for the language, however, dialectal variation is considered an
acceptable and legitimate form of variation in spelling in certain contexts. The
corpus data suggests that as the language has become more endangered since the
1970s, the less that spelling variation is tolerated and used. This needs further
exploration of the sociolinguistic experiences of learning and practising literacy
and the location and analysis of data surrounding literacy rates. Not least, the
theory that even some variability in spelling reduces literacy acquisition should be
questioned. There is also a question of what is a spelling variation and what is a
different lexeme, e.g. are an dèidh, às dèidh, às deòghaidh ('after') considered by
speakers as spelling or lexical variation?
Some participants in the Dlùth is Inneach study explicitly requested that further
progress be made with standardisation of Gaelic spelling. This was particularly
needed for consistency of new terminology and the spelling of loan words:
there's a need to go a bit further with GOC. [...] how we're going tospell loan words to have one piece of advice because, you see loanwords sometimes spelt in three or four different ways. (Bell et al. 2014:B141)
The same observation that '[m]ost respondents felt that the current degree of
confusion and variation on orthography and grammar was unhelpful at best' was
found by Ó Maolalaigh et al. (2009: 30). A corpus analysis can furnish evidence on
which forms have previously been used, what patterns are already at work in the
conventions, and what exceptions are understood, to allow the community to
make informed decisions.
The evidence gathered in this thesis indicates that standardisation in Scottish
Gaelic writing has been a process of gradual, incremental changes. These changes
have taken place within a language ideology that is largely conservative and
inclusive of dialectal variation. Spelling reforms that have broken with the
conservative tendencies have been successful where they have answered
particular needs such as the eighteenth century vernacularisation of the written
212
language. The case studies show that significant orthographic changes in spelling,
such as the removal of the acute and <sd> clusters, can overcome conservative
spelling practices if they accord with the language ideology of the time – in this
case the need for simplification in the sense of fewer spelling variants.The
available, and growing, Scottish Gaelic digital corpus means that data can support
many aspects of linguistic work in historical linguistics, lexicography, and the
creation of language tools to support the language choices of the community.
213
Appendix 1: Corpus Metadata74
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
179 Gairm an De Mhoir(1750)
Richard Baxter,trans. by AlexanderMacFarlane
n/a unknown 1750 Prose Religious 68015 5894 8.67%
171 Ais-éiridh (1751) AlasdairMacMhaighstirAlasdair [AlexanderMacDonald]
c.1698-1770
Moidart 1751 Verse Traditional 24280 7463 30.74%
173 Tiomnadh Nuadh(1767)
Dugald Buchanan,Rev. James Stuart etal.
n/a various 1767 Prose Religious 240774 11217 4.66%
166 LeabharCeasnuigheAithleasuighte(1779)
John Ewart; trans. byDuncan Lothian
Lothian:c.1730
Glen Lyon 1779 Prose Religious 15655 2659 16.98%
169 SaighidearCriosduidh (1797)
Thomas Broughton;trans. anon.
unk. unknown 1797 Prose Religious 9976 2050 20.55%
146 Orain Ghaelach(1801)
Iain Mac Ghrigair[John MacGregor]
unk. Glen Lyon 1801 Verse Traditional 28459 5374 18.88%
145 Searmona (1804) Eobhann MacDiarmaid[Hugh/EwanMacDiarmid]
?-1801 Weem,Perthshire
1804 Prose Religious 80479 5640 7.01%
141 Comhchruinneacha(1813)
various; Paruig Mac-an-Tuairneir [PeterTurner] ed.
n/a various;Cowal
1813 Verse Traditional 69372 10206 14.71%
74 Ordered by year of publication
214
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
135 Co'chruinneachadh(1828)
An t-Urr. TormodMacLeòid 'Caraid nanGàidheal', [Rev.Norman MacLeod]
1783-1862
Morvern(Argyllshire)
1828 Prose/Verse
various 87648 10433 11.90%
128 An Teachdaire ÙrGàidhealach(1835-6)
Lachlan MacLean ed. 1798-1848
various; Coll 1835-6 Prose/Verse
various 129326 15483 11.97%
131 Daoine air anComhairleachadh(1832)
Iain (Eoin)Domhnullach [Rev.Dr John MacDonald]
1779-1849
Caithness 1832 Prose Religious 18083 2276 12.59%
196 Beachd-Chomhairlean(1838)
Sir FrancisMackenzie, trans. byRoderick MacDonald
n/a unknown 1838 Prose Expository 62499 6488 10.38%
125 Sar-Obair namBard Gaelach(1841)
Iain MacCoinnich[John Mackenzie] ed.
1806-1848
various 1841 Verse Traditional 164626 22631 13.75%
105 Laoidhean agusDàin (1868): dàin
Donnachadh Mac-Gilleadhain [Rev.Duncan MacLean]
1795-1871
Killin 1868 Verse Traditional 24372 4664 19.14%
105 Laoidhean agusDàin (1868):Roimh-ràdh
Donnachadh Mac-Gilleadhain [Rev.Duncan MacLean]
1795-1871
Killin 1868 Prose/Verse
Expository 1659 648 39.06%
109 LaoidheanSpioradail (1862)
Daniel Grant unk. Strathspey 1868 Verse Religious 21856 3774 17.27%
100 Am FilidhGaidhealach(1873)
various; HughMacKenzie ed.
n/a various 1873 Verse Literary 21613 5486 25.38%
98 Orain ann saGhailig (1875)
Donnchadh MacCoinnich [DuncanMacKenzie]
unk. Wester Ross(Kinlochewe)
1875 Prose/Verse
7435 2212 29.75%
97 Ordo Missæ(1877)
Catholic Church n/a unknown 1877 Prose Religous 8170 1572 19.24%
215
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
95 An t-Oranaiche(1879)
Gilleasbuig Mac naCeàrdadh [ArchibaldSinclair] ed.
unk. various 1879 Verse Traditional 104321 11721 11.24%
94 Poems and Songs(1880)
Mary MacKellar 1834-1890
Lochaber 1880 Verse Traditional 15445 3859 24.99%
99 The Gaelic Songsof the Late Dr.MacLachlan (1880)
Iain MacLachlainn[Dr. JohnMacLachlan]
1804-1874
Morvern(Rahoy)
1880 Verse Traditional 9859 2867 29.08%
90 Poems (1884) Iain Caimbeul [JohnCampbell]
1823-1897
Oban &Ledaig
1884 Verse Traditional 16525 3470 21.00%
82 Dan Spioradail(1885)
Domhnull Cattanach unk. Kingussie 1885 Verse Religious 2787 889 31.90%
89 Croft Cultivation(1885)
Dr John MacKenzie;trans. by JohnWhyte
unk. Easdale 1885 Prose Expository 3243 923 28.46%
86 Marbh-rainn airDaoine UrramachDiadhaidh (1887)
Uilleam Guinne ('saBhean)
unk. Sutherland 1887 Verse Religious 3182 992 31.18%
83 Dàin agus OrainGhàidhlig (1891)
Mairi Nic-a-Phearsain, Mairi Mhornan Oran, MàiriNighean Iain Bhàin[Mary MacPherson]
1821-1898
Skye 1891 Verse Traditional 46061 6653 14.44%
93 Laithean Ceisde(1880)
unknown; trans. byRev. DuncanMacBeath?
unk. Lewis 1891 Prose Religious 16352 2604 15.92%
80 Dàin Iain Ghobha1 (1893)
Iain Gobha, [JohnMorison]
c.1796-1852
Harris 1893 Verse Religious 47624 7753 16.28%
79 An t-UrramachIain Mac-Rath(1895): Biography
Neacal MacNeacail[Nicol Nicolson]
unk. Lewis 1895 Prose Expository 18211 2788 15.31%
216
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
79 An t-UrramachIain Mac-Rath(1895): Marbhrann
Neacal MacNeacail[Nicol Nicolson]
unk. Lewis 1895 Verse Traditional 1616 631 39.05%
79 An t-UrramachIain Mac-Rath(1895): Roimh-ràdh
Neacal MacNeacail[Nicol Nicolson]
unk. Lewis 1895 Prose Expository 446 190 42.60%
80 Dàin Iain Ghobha2 (1896): Dàin
Iain Gobha, [JohnMorison]
c.1796-1852
Harris 1896 Verse Religious 40000 7119 17.80%
80 Dàin Iain Ghobha2 (1896): Eibhric
Eibhric Morison[Euphemia Morison]
?-1855 Harris 1896 Verse Traditional 2583 1004 38.87%
80 Dàin Iain Ghobha2 (1896): Cliù IainGhobha
Niall Moireastan [NeilMorrison] the PabbayBard
1816-1882
Harris 1896 Verse Traditional 1085 525 48.39%
80 Dàin Iain Ghobha2 (1896):BuanaicheanBhoais
Màiri Mhoireasdan[Mary Morison]
?-1836 Harris 1896 Verse Traditional 447 242 54.14%
76 Na BàirdLeathanach (1898)
various; Rev.Alexander MacLeanSinclair ed.
1840-1924
various; NovaScotia
1898 Verse Traditional 42494 11167 26.28%
77 Leabhar na Ceilidh(1898): Bàrdachd
various; HenryWhyte ed.
1852-1913
Easdale 1898 Verse Traditional 7081 2252 31.80%
77 Leabhar na Ceilidh(1898): Rosg
various; HenryWhyte ed.
1852-1913
Easdale 1898 Prose Traditional 57896 18582 32.10%
78 Leabhar nanGleann (1898)
various; GeorgeHenderson ed.
1866-1912
various 1898 Verse Traditional 71499 14966 20.93%
72 Lòchran an Anma(1906):Urnaighean
Catholic Church n/a unknown 1906 Prose Religious 21179 3196 15.09%
72 Lòchran an Anma(1906): Laoidhean
Catholic Church n/a unknown 1906 Verse Religious 972 467 48.05%
217
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
140 Laoidhean BeanTorra Dhamh(1906)
Bean Torra Dhamh[Mrs Clark, MaryMacPherson]
c.1720-c.1815
Laggan,Badenoch
1906 Verse Religious 4419 1522 34.44%
70 Oiteagan o'n Iar(1908)
Iain MacCormaic[John MacCormick]
c.1870-1947
Mull 1908 Prose Imaginative 32152 7631 23.73%
199 Am Fear-Ciùil(1910)
DòmhnallMacEacharn [DonaldMacKechnie]
1836-1908
Jura 1910 Prose/Verse
Imaginative 48577 6707 13.81%
69 Aig Tigh Na Beinne(1911): Rosg
Catrìona NicIlleBhàinGhrannd [KatharineWhyte Grant]
1845-1928
Learner/Argyll
1911 Prose Expository 3881 1187 30.58%
64 Dùn-Aluinn (1912) Iain MacCormaic[John MacCormick]
c.1870-1947
Mull 1912 Prose Imaginative 49966 5634 11.28%
68 An t-Ogha Mor(1913)
AonghasMacDhonnachaidh[Angus Robertson]
1871-1948
Skye 1913 Prose Imaginative 57432 7933 13.81%
172 Spiritual Songs ofDugald Buchanan(1913)
Dughall Bochanan[Dugald Buchanan]
1716-1768
Strathyre,Perthshire
1913 Verse Religious 10683 2655 24.85%
66 A' BhraisdLathurnach (1914)
EachannMacDhughaill [HectorMacDougal]
1880-1954
Coll 1914 Prose Imaginative 8610 1677 19.48%
62 Cailin Sgiathanach(1923): sgeulachd
Seumas MacLeòid[James MacLeod]
1880-1947
Harris 1923 Prose Imaginative 90349 8542 9.45%
62 Cailin Sgiathanach(1923): roimh-ràdh
Seumas MacLeòid[James MacLeod]
1880-1947
Harris 1923 Prose Expository 608 323 53.13%
57 Litrichean AlasdairMhoir (1932)
Iain N. MacLeòid[John N. MacLeod]
1880-1954
Skye /Lewis 1932 Prose Expository 96238 7398 7.69%
57 Litrichean AlasdairMhoir (1932):Sanas
Dòmhnall S.MacLeòid
1879-1955
Lewis 1932 Prose Expository 568 277 48.77%
218
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
58 Na BairdThirisdeach (1932)
various; EachannCamshron [Rev.Hector Cameron] ed.
1880-1932
Tiree 1932 Verse Traditional 45661 7786 17.05%
182 Highland Songs ofthe Forty-Five(1933)
various; John LorneCampbell ed.
n/a various 1933 Verse Traditional 24373 5610 23.02%
187 Gaelic Songs ofMary MacLeod(1934)
Màiri nigheanAlasdair Ruaidh[Mary MacLeod]; J.Carmichael Watsoned.
17th c. Harris 1934 Verse Traditional 7931 2335 29.44%
56 Bàird Chill-Chomain (1936):DonnchadhMacNimhein
DonnchadhMacNimhein [DuncanMacNiven]
1883-1955
Islay 1936 Verse Traditional 28818 4183 14.52%
56 Bàird Chill-Chomain (1936):TeàrlachMacNimhein
Teàrlach MacNimhein[Charles MacNiven]
1874-1944
Islay 1936 Verse Traditional 3892 1363 35.02%
56 Bàird Chill-Chomain (1936):Roimh-ràdh
NiallMacGilleSheathanaich[Neil Shaw]
1881-1961
Jura 1936 Prose Expository 755 368 48.74%
56 Bàird Chill-Chomain (1936):EoghanMacNìmhein
Niall MacPhail unk. unk. 1936 Verse Traditional 391 215 54.99%
88 Clarsach namBeann (1937):Bàrdachd
Eoghan MacColla[Evan MacColl]
1808-1898
Kenmore,Argyll
1937 Verse Traditional 24511 4559 18.60%
88 Clarsach namBeann (1937): AmBàrd MacColla
Alasdair Friseal [Lt.Col. AlexanderFraser]
unk. unknown 1937 Prose Expository 5280 1389 26.31%
219
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
88 Clarsach namBeann (1937): Abheatha agusObair
Eoghan MacColla &EachannMacDhughaill [EvanMacColl & HectorMacDougal]
n/a various 1937 Prose Expository 2436 764 31.36%
88 Clarsach namBeann (1937):Facal bhon FhearDheasachaidh
EachannMacDhughaill [HectorMacDougal]
1880-1954
Coll 1937 Prose Expository 1002 346 34.53%
54 Am Measg namBodach (1938): AnScarpa
An t-Urr. CalumMacGilleathain
1896-1961
Scarp 1938 Prose Traditional 3190 883 27.68%
54 Am Measg namBodach (1938):Colla
EachannMacDhughaill [HectorMacDougal]
1880-1954
Coll 1938 Prose Traditional 2686 780 29.04%
54 Am Measg namBodach (1938):Muile
Niall Mac 'Ille Mhoire[Neil Morrison]
unk. unk. 1938 Prose Traditional 2742 817 29.80%
54 Am Measg namBodach (1938):Beinn na Bhaoghla
An t-Urr. SeumasMacDhòmhnaill [Rev.James MacDonald]
unk. unk. 1938 Prose Traditional 2273 680 29.92%
54 Am Measg namBodach (1938):Uibhist a Tuath
An t-Urr. NiallMacDhòmhnaill [Rev.Neil MacDonald]
unk. unk. 1938 Prose Traditional 2338 728 31.14%
54 Am Measg namBodach (1938):Uibhist a Deas
DomhnallMacDhòmhnaill[Donald MacDonald]
1912-1989
South Uist 1938 Prose Traditional 2570 802 31.21%
54 Am Measg namBodach (1938):Diura
Niall MacGilleSheathanaich [NeilShaw]
1881-1961
Jura 1938 Prose Traditional 2356 744 31.58%
220
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
54 Am Measg namBodach (1938):Cannaidh, Eige,Ruma
An t-Urr. SomhairleMacIsaac [Rev.Sorley MacIsaac]
unk. unk. 1938 Prose Traditional 2317 736 31.77%
54 Am Measg namBodach (1938):Eisdeal & Luinn
Liusaidh NicCoinnich 1896- unk. 1938 Prose Traditional 2417 770 31.86%
54 Am Measg namBodach (1938): Ant-EileanSgitheanach
An t-Ollamh Niall Ros[Rev. Neil Ross]
1871-1943
Skye 1938 Prose Traditional 2123 717 33.77%
54 Am Measg namBodach (1938):Tiriodh
An t-Urr. EachannCamshron [Rev.Hector Cameron]
1880-1932
Tiree 1938 Prose Traditional 2304 789 34.24%
54 Am Measg namBodach (1938):Eilean I
An t-Urr. CollaDomhnullach
unk. Iona 1938 Prose Traditional 2671 922 34.52%
54 Am Measg namBodach (1938): Ile
Donnchadh MacIain[Duncan Johnston]
1881-1947
Islay 1938 Prose Traditional 2286 790 34.56%
54 Am Measg namBodach (1938):Ratharsair
Iain MacIllEathain[John MacLean]
1909-1970
Raasay 1938 Prose Traditional 1846 654 35.43%
54 Am Measg namBodach (1938):Leòdhas
Seumas MacThomais unk. unk. 1938 Prose Traditional 2312 821 35.51%
54 Am Measg namBodach (1938):Barraidh
Iain Mac'IlleMhaoh unk. unk. 1938 Prose Traditional 1974 742 37.59%
54 Am Measg namBodach (1938): AnCeard Mòr
An t-Urr. CoinneachMacLeòid [Rev.Kenneth MacLeod]
1871-1955
Eigg 1938 Prose Traditional 1450 554 38.21%
221
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
54 Am Measg namBodach (1938): NaHearadh
Seumas MacCoinnich unk. unk. 1938 Prose Traditional 1959 749 38.23%
165 Songs of JohnMacCodrum(1938)
John MacCodrum 1693-1779
North Uist 1938 Verse 19640 4334 22.07%
53 Ban-altrumachdaig an Tigh (1939)
Comhairle Clann anFhraoich
n/a various 1939 Prose Expository 15032 2678 17.82%
52 GriasaicheBhearnaraidh(1940): Bàrdachd
Ailean MacLeòid,'GriasaicheBhearnaraidh', Aileanmac Iain ’icMhurchaidh, [AllanMacLeod]
1858-1939
Harris(Berneray)
1940 Verse Traditional 3492 1063 30.44%
52 GriasaicheBhearnaraidh(1940): Roimh-ràdh
Niall Mac-an-Tuairneir [NeilTurner]
unk. Harris 1940 Prose Expository 912 376 41.23%
52 GriasaicheBhearnaraidh(1940): Oran doAilean MacLeòid
EachannMacFhionghain
1886-1954
Harris(Berneray)
1940 Verse Traditional 408 228 55.88%
52 GriasaicheBhearnaraidh(1940): Facal-toisich
Tòmas M.MacCalmain [Dr.Thomas M.Murchison]
1907-1984
Skye 1940? Prose Expository 212 121 57.08%
51 SgialachdanDhunnchaidh(1944)
Kirkland CameronCraig
?-1965 South Uist 1944 Prose Traditional 38126 3108 8.15%
49 Folksongs andFolklore of SouthUist (1955): Songs
Margaret Fay Shaw 1903-2004
Learner 1955 Verse Traditional 16284 5620 34.51%
222
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
49 Folksongs andFolklore of SouthUist (1955):Proverbs
Margaret Fay Shaw 1903-2004
Learner 1955 Prose Traditional 1315 543 41.29%
49 Folksongs andFolklore of SouthUist (1955):Prayers
Margaret Fay Shaw 1903-2004
Learner 1955 Prose Religious 338 203 60.06%
49 Folksongs andFolklore of SouthUist (1955):Recipes
Margaret Fay Shaw 1903-2004
Learner 1955 Prose Expository 112 71 63.39%
49 Folksongs andFolklore of SouthUist (1955):Riddles
Margaret Fay Shaw 1903-2004
Learner 1955 Prose Traditional 131 86 65.65%
204 Bùrn is Aran(1960):Sgeulachdan
Iain Mac a'Ghobhainn [IainCrichton Smith]
1928-1998
Lewis 1960 Prose Imaginative 15630 1997 12.78%
204 Bùrn is Aran(1960): Bàrdachd
Iain Mac a'Ghobhainn [IainCrichton Smith]
1928-1998
Lewis 1960 Verse Imaginative 2520 925 36.71%
205 An Dubh is AnGorm (1963)
Iain Mac a'Ghobhainn [IainCrichton Smith]
1928-1998
Lewis 1963 Prose Imaginative 36791 3043 8.27%
186 Orain Iain Luim(1964)
Iain Lom [JohnMacDonald]; AnnieM. MacKenzie ed.
c.1625-1707
Keppoch 1964 Verse Traditional 18755 4467 23.82%
47 Sporan Dhòmhnaill(1968)
Dòmhnall Mac an t-Saoir [DonaldMacIntyre]
1889-1964
South Uist 1968 Verse Traditional 64601 9328 14.44%
223
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
45 Dòmhnall RuadhChorùna (1969):Orain
DòmhnallDòmhnallach,Dòmhnall RuadhChoruna [DonaldMacDonald]
1887-1967
North Uist 1969 Verse 15909 4690 29.48%
45 Dòmhnall RuadhChorùna (1969):Dòmhnall Ruadh
Seònaidh A. Mac a'Phearsain [John AlickMacpherson]
1937- North Uist 1969 Prose Expository 834 365 43.76%
45 Dòmhnall RuadhChorùna (1969):Roimh-ràdh
Fred MacAmhlaigh[Fred MacAulay]
1925-2003
North Uist 1969 Prose Expository 241 151 62.66%
74 HebrideanFolksongs 1(1969)
DòmhnallMacCarmaic [DonaldMacCormick]
unk. South Uist 1969 Verse Traditional 13107 2754 21.01%
43 Lus-Chrun aGriomasaidh(1971)
Màiri M.NicGillEathain [MaryM. MacLean]
1921- ? North Uist 1970 Prose Imaginative 41027 4560 11.11%
44 Briseadh naCloiche (1970): AnCluaisean
Fionnlagh MacLeòid[Finlay MacLeod]
unk. Lewis 1970 Prose Imaginative 2743 711 25.92%
44 Briseadh naCloiche (1970):Creic AgusCeannach
Gòrdan Donald[Gordon Donald]
1929- Learner(Tiree)
1970 Prose Imaginative 3044 851 27.96%
44 Briseadh naCloiche (1970): Ant-Ogha
Eilidh Watt 1908-1996
Skye 1970 Prose Imaginative 1651 466 28.23%
44 Briseadh naCloiche (1970):Ozymandias
Anna Chaimbeul[Anna Campbell]
1944- Oban 1970 Prose Imaginative 2685 761 28.34%
224
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
44 Briseadh naCloiche (1970): Sinmar a tha
Pòl MacAonghais[Paul MacInnes]
1928- North Uist 1970 Prose Imaginative 2667 784 29.40%
44 Briseadh naCloiche (1970):Leum NaSinnsearan
Dòmhnall IainMacAoidh [DonaldJohn MacKay]
1930- Harris(Berneray)
1970 Prose Imaginative 3053 899 29.45%
44 Briseadh naCloiche (1970):Mary Kate
CoinneachFionnlasdan[Kenneth Finlayson]
1910- Applecross 1970 Prose Imaginative 1953 589 30.16%
44 Briseadh naCloiche (1970): AnCanna Cocoa
Dòmhnall IainMacDhùghaill[Donald JohnMacDougal]
1921- Barraigh 1970 Prose Imaginative 2066 641 31.03%
44 Briseadh naCloiche (1970): A’Fear A Thainig AirTir
Tormod Caimbeul[Norman Campbell]
1942-2015
Lewis 1970 Prose Imaginative 1883 591 31.39%
44 Briseadh naCloiche (1970):Faoileann
aonghas macneacail 1942- Skye 1970 Prose Imaginative 2327 753 32.36%
44 Briseadh naCloiche (1970):Briseadh naCloiche
Iain Moireach [JohnMurray]
1938- Lewis 1970 Prose Imaginative 1552 608 39.18%
44 Briseadh naCloiche (1970): Nah-Ughdair
Coinneach D.MacDhomhnaill[Kenneth D.MacDonald]
1937- Applecross 1970 Prose Expository 689 278 40.35%
225
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
44 Briseadh naCloiche (1970):Roimh-ràdh
Coinneach D.MacDhomhnaill[Kenneth D.MacDonald]
1937- Applecross 1970 Prose Expository 857 349 40.72%
185 An Clarsair Dall(1970)
An Clarsair Dall[Roderick Morison]
c.1656-c.1713
Lewis 1970 Verse Traditional 6684 1980 29.62%
201 Luach na Saorsa(1970): GuthanBeaga o Latha guLatha
Murchadh Moireach[Murdo Murray]
1890-1964
Lewis 1970 Prose Expository 15717 3033 19.30%
201 Luach na Saorsa(1970):Geographaidh nah-Albainn
Murchadh Moireach[Murdo Murray]
1890-1964
Lewis 1970 Prose Expository 8507 1722 20.24%
201 Luach na Saorsa(1970): Turus donSpainnt
Murchadh Moireach[Murdo Murray]
1890-1964
Lewis 1970 Prose Expository 7125 1709 23.99%
201 Luach na Saorsa(1970): MàiriNighean Iain Bhàin
Murchadh Moireach[Murdo Murray]
1890-1964
Lewis 1970 Prose Expository 6591 1750 26.55%
201 Luach na Saorsa(1970): IainRothach
Murchadh Moireach[Murdo Murray]
1890-1964
Lewis 1970 Prose Expository 3014 984 32.65%
201 Luach na Saorsa(1970): Na h-Orain
Murchadh Moireach[Murdo Murray]
1890-1964
Lewis 1970 Verse Traditional 3920 1354 34.54%
201 Luach na Saorsa(1970): Roimh-Radh
Alasdair I. MacAsgaill[Alex John Macaskill]
1922- ? Lewis 1970 Prose Expository 1403 527 37.56%
42 Tir an Aigh(1971): Stories
Dòmhnall Grannd[Donald Grant]
1903-1970
Skye 1971 Prose Imaginative 23941 7487 31.27%
226
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
42 Tir an Aigh(1971): Gaidhealgu Chul
Tòmas M.MacCalmain [ThomasM. Murchison]
1907-1984
Skye 1971 Prose Expository 606 264 43.56%
42 Tir an Aigh(1971): Poetry
Dòmhnall Grannd[Donald Grant]
1903-1970
Skye 1971 Verse Traditional 5229 2386 45.63%
42 Tir an Aigh(1971): Roimh-ràdh
Iain A.MacDhòmhnaill [JakeA. MacDonald]
1920-1980
Skye 1971 Prose Expository 179 113 63.13%
42 Tir an Aigh(1971): Plays
Dòmhnall Grannd[Donald Grant]
1903-1970
Skye 1971 Prose Imaginative 28578 5187 18.15%
40 A' BhratachDhealrach (1972)
Eilidh Watt 1908-1996
Skye 1972 Prose Imaginative 36917 4301 11.65%
41 Aitealan Dlù isCian (1972)
An t-Urr. CoinneachRos [Rev. KennethRoss]
1914-1990
Skye 1972 Prose Expository 33082 5569 16.83%
46 Saoghal anTreobhaiche(1972)
Aonghas Mac ‘Ill’Fhialain [AngusMacLelllan]
1869-1966
South Uist 1972 Prose Expository 91088 5095 5.59%
183 Silis na Ceapaich(1972)
Sìleas MacDonald;Colm Ó Baoill ed.
c.1660-c.1729
Bohuntin 1972 Verse Traditional 9964 2447 24.56%
35 Suathadh riIomadh Rubha(1973)
Aonghas Caimbeul,'Am Puilean [AngusCampbell]
1903-1982
Lewis 1973 Prose Expository 96267 10167 10.56%
36 Nach Neònach Sin(1973)
Cailein T.MacCoinnich [Rev.Colin N. MacKenzie]
1917-1994
Harris 1973 Prose Expository 20986 3193 15.21%
37 Creach Mhor namFiadh (1973)
Tormod Domhnallach unk. Lewis (Tong) 1973 Prose Imaginative 14268 2555 17.91%
38 Criomagan Ioma-dhathte (1973)
Iain AonghasMacLeòid
1919- Harris 1973 Prose Expository 38650 5002 12.94%
227
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
39 An AghaidhChoimheach(1973)
Iain Moireach [JohnMurray]
1938- Lewis 1973 Prose Imaginative 32832 4671 14.23%
34 Fo Sgail a'Swastika (1974)
Dòmhnull IainMacDhòmhnaill[Donald JohnMacDonald]
1919-1986
South Uist 1974 Prose Expository 23370 3798 16.25%
32 Gàidhlig ann anAlbainn (1976): A'Ghàidhlig agus nah-Ard-Sgoiltean
Murchadh MacLeòid[Murdo MacLeod]
1946- ? Lewis 1976 Prose Expository 3587 728 20.30%
32 Gàidhlig ann anAlbainn (1976): Nah-Oilthighean is naColaisdean
DòmhnallMacAmhlaigh[Donald MacAulay]
1930- Lewis 1976 Prose Expository 3401 774 22.76%
32 Gàidhlig ann anAlbainn (1976):A’Ghàidhlig amBeathafhollaiseach an t-Sluaigh
Dòmhnall IainMacLeòid [DonaldJohn MacLeod]
1943- Harris 1976 Prose Expository 4100 971 23.68%
32 Gàidhlig ann anAlbainn (1976): NaBun-Sgoiltean
Fionnlagh MacLeòid[Finlay MacLeod]
1937- Lewis 1976 Prose Expository 2916 763 26.17%
32 Gàidhlig ann anAlbainn (1976):Leabhraichean,Litreachas,Foillseachadh
RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1976 Prose Expository 2355 628 26.67%
228
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
32 Gàidhlig ann anAlbainn (1976):Craobh-sgaoileadhis Pàipearan
MàrtainnDòmhnallach [MartinMacDonald]
? -2016
Skye 1976 Prose Expository 538 252 46.84%
32 Gàidhlig ann anAlbainn (1976):Dràma
DòmhnallMacGillEathain[Donnie MacLean]
? - 2003 Lewis 1976 Prose Expository 570 271 47.54%
32 Gàidhlig ann anAlbainn (1976):Gàidhlig anAlbainn
RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1976 Prose Expository 481 236 49.06%
32 Gàidhlig ann anAlbainn (1976):Facal sanDealachadh
RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1976 Prose Expository 89 56 62.92%
32 Gàidhlig ann anAlbainn (1976):Roimh-Radh
RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1976 Prose Expository 118 80 67.80%
33 Bith-eòlas (1976) RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1976 Prose Expository 43002 4448 10.34%
198 An t-Aonaran(1976)
Iain Mac a'Ghobhainn [IainCrichton Smith]
1928-1998
Lewis 1976 Prose Imaginative 22495 2319 10.31%
31 Ugam agus Bhuam(1977):Sgeulachdan
Pàdruig Moireasdan 1889-1978
North Uist 1977 Prose Traditional 37687 3222 8.55%
31 Ugam agus Bhuam(1977): Facal onFhearDheasachaidh
Dòmhnall EairdsidhDòmhnallach [DonaldArchie MacDonald]
1929-1999
North Uist 1977 Prose Expository 6373 1253 19.66%
229
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
31 Ugam agus Bhuam(1977): Orain
Pàdruig Moireasdan 1889-1978
North Uist 1977 Verse Traditional 4281 1278 29.85%
31 Ugam agus Bhuam(1977): Roimh-Radh
Pàdruig Moireasdan 1889-1978
North Uist 1977 Prose Expository 233 143 61.37%
31 Ugam agus Bhuam(1977): Oran muChogadh Aifriga
Máiri Dhòmhnaill Òig unk. unknown 1977 Verse Traditional 166 122 73.49%
74 HebrideanFolksongs 2(1977)
various n/a South Uist/Barra/Benbecula
1977 Verse Traditional 19648 3286 16.72%
168 OrainDhonnchaidhBhain (1978)
Donnchadh Ban Macan t-Saoir [DuncanBan MacIntyre]
1724-1812
Glenorchy 1978 Bàrdachd 37160 6439 17.33%
30 Deireadh anFhoghair (1979)
Tormod Caimbeul'Tormod a' Bhocsair'[Norman Campbell]
1942-2015
Lewis 1979 Prose Imaginative 33601 4180 12.44%
191 Eachann Bacach(1979)
various, Colm ÓBaoill ed.
n/a various 1979 Verse Traditional 13383 5642 42.16%
26 Oighreachd isGabhaltas (1980):Aramach amBearnaraigh 1874
DòmhnallMacAmhlaigh[Donald MacAulay]
1930- Lewis 1980 Prose Expository 5640 1101 19.52%
26 Oighreachd isGabhaltas (1980):Blar a’ Chumhaing
Iain A.MacDhòmhnaill [JakeA. MacDonald]
1920-1980
Skye 1980 Prose Expository 5391 1074 19.92%
26 Oighreachd isGabhaltas (1980):Aimhreit anFhearainn anTiriodh 1886
Dòmhnall Meek[Donald E. Meek]
1949- Tiree 1980 Prose Expository 4834 1007 20.83%
230
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
26 Oighreachd isGabhaltas (1980):Mar a Ghabh naDaoineBhatarsaigh
Lisa Storey 1935- Barra(Vatersay)
1980 Prose Expository 3541 765 21.60%
26 Oighreachd isGabhaltas (1980):Reud na Pairce
Iain M. MacLeòid[John M. MacLeod]
1924- Lewis 1980 Prose Expository 5306 1261 23.77%
26 Oighreachd isGabhaltas (1980):Toirt a-machBhaile Raghnaill
UilleamMacDhòmhnaill[William MacDonald]
1949- North Uist 1980 Prose Expository 2636 716 27.16%
26 Oighreachd isGabhaltas (1980):Aimhreit Aignis1888
Iain MacArtair [JohnMacArthur]
1892-1980
Lewis 1980 Prose Expository 2529 810 32.03%
26 Oighreachd isGabhaltas (1980):Roimh-ràdh
DòmhnallMacAmhlaigh[Donald MacAulay]
1930- Lewis 1980 Prose Expository 603 278 46.10%
74 HebrideanFolksongs 3(1981)
various n/a SouthUist/Barra/Benbecula
1981 Verse Traditional 17641 3071 17.41%
28 Air Mo Chuairt(1982)
Ealasaid Chaimbeul 1913- Barra 1982 Prose Expository 44415 4804 10.82%
28 Air Mo Chuairt(1982): còmhdach
unknown n/a n/a 1982 Prose Expository 107 65 60.75%
29 Creachadh naClàrsaich (1982)
RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1982 Poetry Imaginative 30464 6194 20.33%
27 Suileabhan (1983) CalumMacFhearghuis[Calum Ferguson]
1929- Lewis 1983 Prose Expository 47297 5362 11.34%
231
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
27 Suileabhan(1983):Naidheachd
Eòghainn MhurchaidhEòghainn
unk. Lewis 1983 Prose Traditional 1007 379 37.64%
27 Suileabhan(1983): AnnasNaidheachd
CalumMacFhearghuis[Calum Ferguson]
1929- Lewis 1983 Prose Expository 973 379 38.95%
27 Suileabhan(1983): Roimh-ràdh
CalumMacFhearghuis[Calum Ferguson]
1929- Lewis 1983 Prose Expository 1399 603 43.10%
22 Eòin an Àite(1986)
Frang U. Rennie[Frank U. Rennie]
unk. unknown 1986 Prose Expository 1979 354 17.89%
23 Seann TaigheanTirisdeach (1986):main text
Ailean Boyd unk. Tiree 1986 Prose Expository 6900 1548 22.43%
23 Seann TaigheanTirisdeach (1986):Na Daoine
Ailean Boyd unk. Tiree 1986 Prose Expository 381 126 33.07%
23 Seann TaigheanTirisdeach (1986):Ro-ràdh
Ailig MacArtair unk. Tiree 1986 Prose Expository 395 194 49.11%
19 Air Druim an EichSgiathaich (1987)
Pòl Mac a'Bhreatunnaich
1923- ? Barra 1987 Prose Expository 32292 3552 11.00%
20 A' Ghàidhlig annsan EileanSgitheanach(1987)
Comunn na Gàidhlig n/a unknown 1987 Prose Expository 2789 620 22.23%
232
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
17 Loch Druidibeg(1988)
NicLeòid, C.;NicLeòid, F.;NicDhomhnaill, A.[ComhairleGleidhteachaisNaduir/ NatureConservancy Council]
n/a various 1988 Prose Expository 1354 487 35.97%
18 Dealbh-chruth nanEilean Siar (1988)
Comhairle nan Eilean[Western IslesCouncil]
n/a unknown 1988 Prose Expository 8429 1432 16.99%
48 SgrìobhaidheanChoinnichMhicLeòid (1988)
An t-Urr. CoinneachMacLeòid [Rev.Kenneth MacLeod]
1871-1955
Eigg 1988 Prose Expository 65109 18885 29.01%
16 Spuirean na h-Iolaire (1989)
Iain Macleòid [JohnMacLeod]
1933- ? Skye 1989 Prose Imaginative 34800 3234 9.29%
14 A' Sireadh anSgadain (1990):sgeulachd
Calum MacMhaoilein[MacMillan]
unk. Lewis 1990 Prose Imaginative 32019 2701 8.44%
14 A' Sireadh anSgadain (1990):Roimh-ràdh
Calum MacMhaoilein[MacMillan]
unk. Lewis 1990 Prose Expository 479 217 45.30%
15 Bàrdachd naRoinn-Eòrpa(1990): CrichtonSmith
Iain Mac a'Ghobhainn [IainCrichton Smith]
1928-1998
Lewis 1990 Verse Imaginative 4173 1327 31.80%
15 Bàrdachd naRoinn-Eòrpa(1990): Whyte
Crìsdean Whyte[Christopher Whyte]
1952- Learner(Skye)
1990 Verse Imaginative 4452 1604 36.03%
15 Bàrdachd naRoinn-Eòrpa(1990):MacThòmais
RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1990 Verse Imaginative 3627 1321 36.42%
233
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
15 Bàrdachd naRoinn-Eòrpa(1990): BàrdachdGhalicia
Niall A. R. MacAoidh unk. unk. 1990 Verse Imaginative 392 174 44.39%
15 Bàrdachd naRoinn-Eòrpa(1990): Neruda
FearghasMacFhionnlaigh
1948- Learner 1990 Verse Imaginative 1513 687 45.41%
15 Bàrdachd naRoinn-Eòrpa(1990): Facal-Tòisich
RuaraidhMacThòmais [DerickS. Thomson]
1921-2012
Lewis 1990 Prose Expository 454 228 50.22%
15 Bàrdachd naRòinn-Eòrpa(1990): Per JakezHelias
Anna Frater [AnneFrater]
1967- Lewis 1990 Verse Imaginative 451 234 51.88%
15 Bàrdachd naRoinn-Eòrpa(1990): Neill
Uilleam Neill [WilliamNeill]
1922- Learner 1990 Verse Imaginative 1933 1032 53.39%
15 Bàrdachd naRoinn-Eòrpa(1990): Brecht
[Brecht] IainMacDhòmhnaill &Mairead Ryan [IainMacDonald &Margaret Ryan]
unk. various 1990 Verse Imaginative 398 214 53.77%
15 Bàrdachd naRoinn-Eòrpa:Bàrdachd Fhraingis
AlasdairMac’IlleMhaoil
unk. unknown 1990 Verse Imaginative 543 297 54.70%
15 Bàrdachd naRoinn-Eòrpa(1990): King
Dennis King unk. learner(california)
1990 Verse Imaginative 422 234 55.45%
15 Bàrdachd naRoinn-Eòrpa(1990): Horace
Iain MacIllEathain[John MacLean]
1909-1970
Raasay 1990 Verse Imaginative 485 272 56.08%
234
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
15 Bàrdachd naRoinn-Eòrpa(1990): A’ bhanais
Fionnlagh IainMacDhòmhnaill
1926- Harris 1990 Verse Imaginative 265 149 56.23%
15 Bàrdachd naRoinn-Eòrpa(1990): Andreus
Maoilios Caimbeul[Myles M. Campbell]
1944- Skye 1990 Verse Imaginative 281 162 57.65%
15 Bàrdachd naRòinn-Eòrpa(1990): UrnuighUilleim Chràbhaich
Cailein T.MacCoinnich [Rev.Colin N. MacKenzie]
1917-1994
Harris 1990 Verse Imaginative 679 395 58.17%
15 Bàrdachd naRoinn-Eòrpa(1990): CampbellHay
Deòrsa Caimbeul Hay[George CampbellHay]
1915-1984
Learner 1990 Verse Imaginative 1632 954 58.46%
15 Bàrdachd naRoinn-Eòrpa(1990):Shakespeare
Uisdean Laing [HughLaing]
1889-1974
South Uist 1990 Verse Imaginative 492 289 58.74%
15 Bàrdachd naRoinn-Eòrpa(1990): DoEòghannMacLachlainn
Iain Stoddart unk. unk. 1990 Verse Imaginative 239 141 59.00%
15 Bàrdachd naRoinn-Eòrpa(1990):Shakespeare
Iain MacIllEathain[John MacLean]
1909-1970
Raasay 1990 Verse Imaginative 319 200 62.70%
15 Bàrdachd naRoinn-Eòrpa(1990): Meadhonan Iomaill
Richard Cox 1954- Learner 1990 Verse Imaginative 140 88 62.86%
235
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
15 Bàrdachd naRoinn-Eòrpa(1990): MiCuimhneachadh
Raghnall Moireasdan unk. unk. 1990 Verse Imaginative 277 177 63.90%
15 Bàrdachd naRoinn-Eòrpa(1990): A’ fuireach’s a' falbh
Iain MacLeòid unk. unk. 1990 Verse Imaginative 133 87 65.41%
15 Bàrdachd naRoinn-Eòrpa(1990): NuairShiùbhlas Mi
Tormod MacLeòid'Am Bàrd Bochd'[Norman MacLeod]
1904-1968
Lewis 1990 Verse Imaginative 138 91 65.94%
15 Bàrdachd naRoinn-Eòrpa(1990): ÒranBrosnachaidh
Tormod Burns[Norman Burns]
unk. Learner 1990 Verse Imaginative 101 68 67.33%
15 Bàrdachd naRoinn-Eòrpa(1990): Meadhonna Beatha
Victor Price 1930- Learner(Lewis)
1990 Verse Imaginative 102 69 67.65%
15 Bàrdachd naRoinn-Eòrpa(1990):Dìomhanas
Coinneach Ros [Rev.Kenneth Ross]
1914-1990
Skye 1990 Verse Imaginative 137 97 70.80%
15 Bàrdachd naRoinn-Eòrpa(1990): GarbhanMacAoidh
Garbhan MacAoidh[Girvan McKay]
1929- Learner 1990 Verse Imaginative 647 375 57.96%
50 An NeamhnaidLuachmhor(1990): Laoidhean
EachannMacFhionghain
1886-1954
Harris(Berneray)
1990 Verse Traditional 44799 4283 9.56%
236
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
50 An NeamhnaidLuachmhor(1990): Facal-toisich
An t-Urr. AonghasMacPhàrlain [Rev.Angus MacFarlane]
unk. unknown 1990 Prose Expository 691 319 46.16%
50 An NeamhnaidLuachmhor(1990): Ionndrain
Murchadh MacSuain unk. Harris(Berneray)
1990 Verse Traditional 635 327 51.50%
50 An NeamhnaidLuachmhor(1990): Laoidh dodh’Eachann
An t-Urr. UilleamMacGhillFhinnein[Rev. WilliamMacLennan]
unk. unknown 1990 Verse Traditional 465 254 54.62%
50 An NeamhnaidLuachmhor(1990): Marbhranndo dh’Eachann
RuairidhMacFhionghain
unk. Harris(Berneray)
1990 Verse Traditional 400 219 54.75%
50 An NeamhnaidLuachmhor(1990): Marbhrann
Dòmhnall Màrtainn unk. unknown 1990 Verse Traditional 386 214 55.44%
50 An NeamhnaidLuachmhor(1990): còmhdach
unknown unk. unknown 1990 Prose Expository 213 125 58.69%
50 An NeamhnaidLuachmhor(1990): Marbhrannair EachannMacFhionghain
RuairidhMacFhionghain
unk. Harris(Berneray)
1990 Verse Traditional 356 210 58.99%
12 Croitearachd(1991)
Museum nan Eilean n/a unknown 1991 Prose Expository 249 138 55.42%
13 Iasgach (1991) Museum nan Eilean n/a unknown 1991 Prose Expository 394 208 52.79%11 Am Fear
Meadhanach(1992)
Alasdair Caimbeul,'Alasdair a' Bhocsair'[Alasdair Campbell]
1941- Lewis (Ness) 1992 Prose Imaginative 43702 5223 11.95%
237
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
10 Coimhead air anTaigh-Dubh annan Arnol (1994)
RuaraidhMacIlleathain [RoddyMacLean]
1954 - Learner(Skye)
1994 Prose Expository 2559 795 31.07%
8 Hiort (1995): Rosg CalumMacFhearghuis[Calum Ferguson]
1929- Lewis 1995 Prose Expository 57235 6422 11.22%
8 Hiort (1995):Bàrdachd
various; CalumMacFhearghuis[Calum Ferguson]ed.
1929- Lewis 1995 Verse Traditional 7091 3420 48.23%
84 Tuath is Tighearna(1995)
various; Donald Meeked.
n/a various 1995 Verse Traditional 16315 3729 22.86%
174 Duanaire Colach(1997): intro
Colm Ó Baoill 1938- learner 1997 Prose Expository 18663 3262 17.48%
174 Duanaire Colach(1997): notes
Colm Ó Baoill 1938- learner 1997 Prose Expository 17577 3090 17.58%
174 Duanaire Colach(1997): bàrdachd
various; Colm ÓBaoill ed.
n/a Coll 1997 Verse Literary 14790 6447 43.59%
7 Moch is Anmoch(1998): Donald A.MacNeill
Donald A. MacNeil unk. Colonsay 1998 Verse Traditional 1900 756 39.79%
7 Moch is Anmoch(1998): MurchadhRuadh
Folalaidh NicNèill[Flora Margaret'Folalie' McNeill]
unk. Colonsay 1998 Verse Traditional 320 186 58.13%
7 Moch is Anmoch(1998): An t-Slat-iasgaich
Alexander Darroch unk. Colonsay 1998 Verse Traditional 232 138 59.48%
7 Moch is Anmoch(1998): An t-Achmhasan
unknown unk. Colonsay 1998 Verse Traditional 163 116 71.17%
238
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
4 BàrdachdDhòmhnaillAlasdair (1999):Bàrdachd
Dòmhnall AlasdairDòmhnallach [DonaldAlasdair MacDonald]
1919-2003
Lewis 1999 Verse Traditional 14047 3054 21.74%
4 BàrdachdDhòmhnaillAlasdair (1999):ro-ràdh
Dòmhnall AlasdairDòmhnallach [DonaldAlasdair MacDonald]
1919-2003
Lewis 1999 Prose Expository 2162 690 31.91%
4 BàrdachdDhòmhnaillAlasdair (1999):Facal-toisich
Joan Dhòmhnallach[Joan MacDonald]
unk. Lewis 1999 Prose Expository 677 316 46.68%
3 Smuaintean foÉiseabhal (2000)
Dòmhnall AonghaisBhàin [DonaldMacDonald]
1926-2000
South Uist 2000 Verse Traditional 15451 3404 22.03%
3 Smuaintean foÉiseabhal (2000):Am Bàrd
Ailig O Hianlaidh[Alex O'Henley]
1965- South Uist 2000 Prose Expository 1496 533 35.63%
3 Smuaintean foÉiseabhal (2000):Roimh-ràdh
Aonghas PàdraigCaimbeul [AngusPeter Campbell]
1954- South Uist 2000 Prose Expository 1378 597 43.32%
2 SgeulachdanDhòmhnaillAlasdair (2001)
Dòmhnall AlasdairDòmhnallach [DonaldAlasdair MacDonald]
1919-2003
Lewis 2001 Prose Traditional 34035 3700 10.87%
2 SgeulachdanDhòmhnaillAlasdair (2001):còmhdach
unknown unk. unknown 2001 Prose Expository 159 99 62.26%
1 Dàin do Eimhir(2002)
Somhairle MacGill-Eain [SorleyMacLean]
1911-1996
Raasay 2002 Verse Imaginative 6811 2098 30.80%
239
DASGno.
Text Author Authordates
Dialect Year ofPublication
Type Genre Tokens Types TTR
203 Tocasaid 'Ain Tuirc(2004)
DonnchadhMacGilliosa
1941- Lewis 2004 Prose Imaginative 33024 5028 15.23%
202 Na Klondykers(2005)
Iain FionnlaghMacLeòid [Iain F.MacLeod]
1973- Lewis 2005 Prose Imaginative 68355 4694 6.87%
197 Saoghal Bana-mharaiche (2007)
Seosamh Watson ed. n/a Easter Ross 2007 Prose Traditional 72767 7015 9.64%
240
Appendix 2: DASG metadata example
Digital Archive of Scottish GaelicMetadata for texts common to Corpas na Gàidhlig and Faclair na Gàidhlig have been provided bythe Faclair na Gàidhlig project. We are very happy to acknowledge here Dr Catriona Mackie’ssterling work in producing this data; the University of Edinburgh for giving us permission to use andpublish the data; and the Leverhulme Trust whose financial support enabled the production of themetadata in the first place. The metadata is provided here in draft form as a useful resource forusers of Corpas na Gàidhlig. The data is currently being edited and will be updated in due course.
Metadata © University of Edinburgh
Metadata for text 199
No. words in text 108250
Title Am Fear-Ciuil. Dain, Orain, Oraidean, is Sgeulachdan.
Author MacEacharn, Domhnull
Editor Mac Fhionghain, Domhnall
Date Of Edition 1910
Date Of Language 1900-1949
Publisher John Grant
Place Published Edinburgh
Volume N/A
Location National and academic libraries
Geographical Origins Jura
Register Literature, Prose and Verse
Reference Style
Alternative Author Name MacKechnie, Donald
Manuscript Or Edition Ed.
Size And Condition 20cm x 14cm
Short Title Am Fear-Ciuil
Reference Details EUL, Scottish Studies Library: G4(G)MacK
Number Of Pages xvi, 336
Gaelic Text By N/A
241
Illustrator N/A
Social Context Donald MacKechnie was born in Glengarisdale, on the shore ofCorryvreckan in the north of Jura on 25th December 1838. Therewas no school nearby, but Donald was taught to read and write ata school near where his maternal grandfather lived. He had anactive life as a boy, spending much time outdoors, shooting andfishing. MacKechnie moved to Glasgow while a young man. Heattended evening classes to improve his education, and began toread widely, particularly English literature. He eventually madehis home in Edinburgh, marrying Elizabeth Jane Sutherland thereon 5th June 1868. They had seven children. Living in Edinburgh,MacKechnie was part of the Gaelic literati which, at that time,included Alexander Carmichael, Donald MacKinnon, and SheriffNicolson. MacKechnie mentions such gatherings in his song 'A'Chéilidh' (pp. 93-97). MacKenchnie was troubled by illnessthroughout his adult life. He died in Edinburgh in May, 1908.MacKechnie wrote both poetry and prose, and he contributed anumber of works to An Gàidheal, and other journals, under thename 'Am Bard Luideagach'. MacKechnie won a number of prizesat the National Mòd, including first prize for his poem 'Am Fear-Ciùil' at the Oban Mòd (in 1892?). He also translated a number ofpoems and songs into Gaelic, including verse from OmarKhayyám's Rubaiyat. However, MacKenzie is best known for hisprose writings, and particularly for those essays in which hediscusses man's relationship with animals. In his obituary in TheCeltic Review, Donald MacKinnon noted that while MacKechnie'spoetry was perhaps not quite as good as that of some of hiscontemporaries, 'I do not know that any Gaelic writer, of moderntimes at any rate, excels him in prose. ... Here the highly trainedintellect of a very capable man gives his own views of men andthings, with a probing and questioning almost Socratic in itspatience and persistence, and with a terseness and crispness ofphrase more akin to French than to Gaelic prose' (MacKinnon, p.94).
Contents This volume begins with a Clar-Innsidh (pp. ix-xi), followed by anaddress Do'n Leuighadair by the author (pp. xii-xiv), followed byAn Roimh-radh (pp. xv-xvi) in verse form. There is a photographof the author opposite the Gaelic title page. This (second)edition contains 13 poems; 18 songs; 6 translations; and 19 proseitems. Verse The poems and songs in this volume cover a varietyof subjects, and are all fairly light-hearted in nature. None areparticularly noteworthy, and few of them are popular today.Subjects covered include nature, e.g. 'Guth a' chuain' (pp. 9-11)and 'Seachran Seilg' (pp. 26-28), both of which also containelements of religion; love, e.g. 'An Ribhinn Òg D'an D'thug MiGràdh' (p. 74) and 'Bean a' Chòtain Ruaidh' (pp. 81-82); war, e.g.the four poems that make up 'An Cogadh an Africa-mu-dheas' (pp.61-66); and sense of place, e.g. in 'Chunnaic mi na Gruagaichean'(pp. 67-68), 'Cuairt 'san Fhrìdh' (pp. 75-76), and 'Am Bothan Beag'(pp. 79-80), which also touch on the changes that have takenplace in the Highlands during the author's lifetime. His knowledgeof world history and geography is apparent in 'ImpireachdBhreatuinn' (pp. 54-58) which comprises a conversation betweenAm Bard and A' Ghrian. The translations include an Englishtranslation of MacKechnie's 'Am Fear-Ciùil'; and translations intoGaelic of Thomas Campbell's 'The Soldier's Dream', Robert Burns's'My Love is Like a Red, Red Rose', and verses from OmarKhayyám's Rubaiyat. Prose MacKechnie is particularly well-knownfor his humorous stories about his relationship with variousanimals. In these stories, MacKechnie ruminates on what itmeans to be human, and ponders whether the animals he writes
242
about are any different from us in the way they behave. Many ofthese stories feature his cat (Tómas) and his dog (Yarrow), eachof whom have a story dedicated to them ('An Cat', pp. 126-34,and 'An Cù', pp. 135-41). The best known of these works isprobably 'Am Fiadh' (pp. 292-319), which first appeared in printin the second edition of 1910. MacKechnie's stories are generallywritten in the first person, as the author relates the tale ofsomething that happened to him. His wife occasionally makes anappearance in the stories, usually in an antagonistic role, e.g. in'Ath-leasachadh' (pp. 155-63) and at the beginning of 'An Seangan'(p. 149-54). MacKechnie also frequently makes use of thecontrast between the ordinary people (such as himself) and theupper classes, e.g. in 'Am Fiadh' (where the new laird's wife isterrified of the deer which MacKechnie, as a boy, was chargedwith looking after) and in 'An Seangan' (where a shambolic sceneon Salisbury Craggs, in which the author is assailed by a colony ofants, is witnessed by an upper-class young lady, much to herastonishment). MacKechnie also wrote some factual prose,including a talk on 'Carmina Gadelica' (pp. 164-86), which wasread before the Jura Association in Glasgow, and an essay on'Omar Khayyám' (pp. 200-16), which included translations of someof his verses. In 'Còmhradh', we are given the content of adialogue between Eoghann Og agus Eachann Ruadh, on thesubject of the Eachann's command of the English language. Twoof the essays are written in the form of letters: 'Tannaisg annLaithean a dh'Fhalbh' (pp. 252-60) and 'Litir thun an "Deo-Greine"'(pp. 320-27). 'Tannaisg ann Laithean a dh'Fhalbh' (pp. 252-60) and'Gaisgeach na Sgeithe Deirge' (pp. 261-82) look at legends andfolktales.
Sources
Language The strengths of this volume lie in MacKechnie's prose writings.The author has a wonderful turn of phrase and uses richidiomatic Gaelic. This can be seen, for example, at the beginningof his essay on 'An Cat': Theirinn ri caraid no bana-charaid 's ambith a thig trasd air an duilleig so gun mhor-shuim a ghabhail dhi,nach 'eil dad innte 's fhiach a leughadh. Cha 'n 'eil mi, mar gu 'mb' eadh, ach 'g am thoileachadh fhein, 's an uair a tha duine 'g athoileachadh fein faodaidh e bhi cinnteach nach 'eil etoileachadh muinntir eile' (p. 126). MacKechnie goes on to give usan example of this in relation to his cat. There is muchterminology of interest in MacKechnie's prose writings,particularly with regard to the author's turn of phrase. Examplesinclude 'gu robh an tigh r'a theinidh' (p. 132), 'biomaid a' bogadhnan gad' (p. 233), 'leumadh mo chride-sa as a chochull' (p. 238),'a' toirt sràid do 'n chuilein mhadaidh agam' (p. 188), 'Mo ghillegeal!' (p. 189), 'ann am chnap-starra' (p. 189), 'maol-cheannach'(p. 189), ''san odhar-dhorcha' (p. 255), and 'eadar fheala-dhà 'sda-rìreadh' (p. 283). 'Tannaisg ann Laithean a dh'Fhalbh' (pp. 252-60) and 'Gaisgeach na Sgeithe Deirge' (pp. 261-82) contain someterminology related to legends and folktales, e.g. 'Mac anEarraidh Uaine, no Chochaill Uaine' (p. 280), 'Gruagach nanCumha' (p. 280), 'Ridir a' Chuirn agus Ridir a' Chlaidhimh' (p 280),'Buidseach Endor, 's air Tàillear na Manachainn' (p. 253),'Lachlann Mor 's Dubh-sìth' (p. 253), and 'Aoradh Aithrichean, noAoradh Thannasg' (p. 258). MacKechnie's verse contains littleterminology of note.
Orthography The author's dialect might be evident in the use of terms such as'gabhail iolla riù' (An Cat, p. 117), 'air t' aghairt' (An Cat, p. 120),'thun' (An Cat, p. 120), 'ruais' (p. 221), 'ciogailteach' (Còmhradh,p. 175), and 'trasd air' (An Còmhradh, p. 175). The orthography is
243
that of 1910, and has been modernised from the 1904 edition(see Editions below). Domhnall Mac Fhionghain notes, at thebeginning of the volume, 'Dh' ullaich an t-Ughdar caomh an daraclo-bhualadh de'n "Fhear-Chiuil" goirid mu 'n d'thainig a' chriochair. Tha an leabhar air a chur a mach mar dh'fhag e e, ach amhain gu bheil, a nis 's a rithist, geagan atharrachaidh 's anlitreachadh, agus roinn de mhearachdan a' chlò air anceartachadh'.
Other Sources
Further Reading MacKinnon, Donald 'The Late Mr. Donald MacKechnie', The CelticReview 5, 1908-09, pp. 92-96.
Powered by CQPWeb
244
Bibliography
All author names, places and publishers are given as printed. Transliterations into
English are given in square brackets where appropriate. All M', Mc and Mac
surnames are treated as Mac for the alphabetical listings. Spaces or hyphens such
as Mac a' Ghobhainn or Mac-Mhuirich are also disregarded.
Corpus Texts
An Comunn Gàidhealach. 1938. Am Measg Nam Bodach (Glasgow: An ComunnGàidhealach)
Baxter, Richard. 1750. Gairm an De Mhoir do 'n t Sluagh Neimh-Iompoichte, Iompochadhagus Bith Beo le Richard Baxter. Eidir-theangaicht' o Ghaill-bhearla chumGaoidheilg Albannaich, air Iartas agus Costas Eirionnaich Dheagh-runaich oirdheirc,chum leass coitcheann Gaoidhealtachd Alba (Glassacha [Glasgow]: Clo-bhuailt' annGlassacha le Roib. agus Aind. Foulis Cloi-fheara an Oill-tigh [Robert & AndrewFoulis, Printers to the University])
Boyd, Ailean. 1986. Seann Taighean Tirisdeach (Tiree: Cairdean nan Taighean Tugha)
Breeze Jones, E. V. 1986. Eòin an Aite (Steòrnabhagh [Stornoway]: Acair)
Broughton, Thomas. 1797. Saighidear Criosduidh: no an Dleasnais Iomchuidh chaumBeatha Dhiadhaidh Chaithe, air an Sparradh air an Armailt (Dun-Eaduin[Edinburgh]: Eoin Moir)
Buchanan, Dugald (ed.). 1767. Tiomnadh Nuadh ar Tighearna agus ar Slanuigh-Fhir IosCriosd. Eidir-theangaicht' o'n Ghreugais chum Gaidhlig Albannaich. Maille reSeòlannaibh Aith-ghearra chum a' Chán'ain sin a Leughadh (Dun-Eudain [Edinburgh]:Balfour, Auld, & Smellie)
Buchanan Dugald. 1913. The Spiritual Songs of Dugald Buchanan, ed. by Rev. DonaldMacLean (Edinburgh: John Grant)
Caimbeul, Alasdair. 1992. Am Fear Meadhanach (Ross-shire: Druim Fraoich)
Caimbeul, Aonghas. 1973. Suathadh Ri Iomadh Rubha, ed. by Iain Moireach (Glaschu[Glasgow]: Gairm)
Caimbeul, Tormod. 1979. Deireadh an Fhoghair (Edinburgh: W. and R. Chambers)
Campbell, John. 1884. Poems (Edinburgh: Maclachlan and Stewart)
Campbell, John Lorne (ed.). 1933. Highland Songs of the Forty-Five (Edinburgh: JohnGrant)
Campbell, John Lorne (ed.). 1969. Hebridean Folksongs, A Collection of Waulking Songsby Donald MacCormick in Kilphedir in South Uist in the year 1893 (Oxford:Clarendon Press)
Campbell, John Lorne (ed.). 1977. Hebridean Folksongs 2: Waulking songs from Barra,South Uist, Eriskay and Benbecula (Oxford: Clarendon Press)
245
Campbell, John Lorne (ed.). 1981. Hebridean Folksongs 3: Waulking songs fromVatersay, Barra, Eriskay, South Uist and Benbecula (Oxford: Clarendon Press)
Camshron, an t-Urr. Eachann [Cameron, Rev. Hector] (ed.). 1932. Na Baird Thirisdeach,Saothair Ar Co-luchd-duthcha aig an Tigh 's Bho'n Tigh (Tiree: A' ChomunnThirisdeach [The Tiree Association])
Catholic Church. 1906. Lòchran an Anma: Leabhar-ùrnaigh Caitliceach (Edinburgh:Sands agus A' Chuideachd)
Catholic Church. 1877. Ordo Missæ, 'an Laidinn 's 'an Gàidhlig. Agus ùrnaighean eile(Edinburgh: Lorimer & Gillies)
Cattanach, Domhnull. 1885. Dan Spioradail le Domhnull Cattanach, Sliabh SgirCeannghuisich (Inverness: Printed at the "Northern Chronicle" Office)
Chaimbeul, Ealasaid. 1982. Air Mo Chuairt (Stornoway: Acair)
Cham'ron, Marairead. 1785. Orain Nuadh Ghaidhealach (Clodh-bhuailt' ann Dun-Eidin[Edinburgh]: le D. Mac-Phatraic, airson an Ughdair [by D. MacPatrick for theauthor])
Clark, Mary. 1902. Laoidhean Bean Torra Dhamh, ed. by Rev. Thomas Sinton (Inverness:Northern Chronicle Office)
Comhairle Clann an Fhraoich (ed.). 1939. Ban-Altrumachd aig an Tigh (Glasgow: AnComunn Gàidhealach)
Comhairle Glèidhteachais Nàdur. 1988. Loch Druidibeg, Comraich Naiseanta Naduir(Inverness: Comhairle Glèidhteachais Nàdur (Nature Conservancy Council)
Comhairle nan Eilean. 1988. Dealbh-chruth nan Eilean Siar: aithisg sgrìobhte (WesternIsles Structure Plan: written statement) (Stornoway: Comhairle nan Eilean)
Comunn na Gàidhlig. 1987. A' Ghaidhlig anns an Eilean Sgitheanach Plana LeasachaidhDeich Bliadhna (Gaelic in Skye: A Ten Year Development Plan) (Inverness: Comunnna Gàidhlig)
Craig, K. C. (ed.). 1944. Sgialachdan Dhunnchaidh: seann sgialachdan air an gabhail leDunnchaidh Mac Dhomhnaill Ac Dhunnchaidh, Uibhist a Deas, mar a chual e aigathair fhein iad (Glasgow: Alasdair Matheson & Co.)
Dòmhnallach, Dòmhnall. 1969. Dòmhnall Ruadh Chorùna: òrain is dàin, ed. by FredMacAmhlaigh (Glasgow: Gairm)
Dòmhnallach, Dòmhnall Alasdair. 1999. Bàrdachd Dhòmhnaill Alasdair (Stornoway:Acair)
Dòmhnallach, Dòmhnall Alasdair. 2001. Sgeulachdan Dhòmhnaill Alasdair (Stornoway:Acair)
Domhnullach, Eoin. 1832. Daoine air an Comhairleachadh an aghaidh bhi DeanamhCroin orra fhein. Searmoin, a Thugadh Seachad an' Inerpheafaran aig an am an doBhris an Galar d'an Goirear an Colera Mach sa Bheile (Inbherneis [Inverness]:Daibhidh Aitcinn [David Aitken])
Domhnallach, Tormod. 1973. Creach Mhor nam Fiadh (Stornoway: Stornoway Gazette)
246
Ewart, John. 1779. Leabhar Ceasnuighe Aithleasuighte, ata nochda MearachdaSonruighte Eaglais na Roimh, trans. by Duncan Lothian (Duneudain [Edinburgh]:Daibhidh Mac-Phatric)
Grannd, Dòmhnall. 1971. Tir an Aigh, ed. by Iain A. MacDhomhnaill (Glasgow: Gairm)
Grant, Daniel. 1862. Laoidhean Spioradail (Perth: J. Dewar; Edinburgh: D.R. Collie;Elgin: W. M'Donald)
Grant, Katharine Whyte. 1911. Aig Tigh na Beinne (Oban: Hugh MacDonald; Glasgow:Alex. M'Laren & Son)
Guinne, Uilleam, & a Bhean. 1887. Marbh-rainn air daoine urramach diadhaidh a bha 'anSgire Dhornaich, 'an Cataobh (Dingwall: Printed at the Ross-shire Journal Office)
Henderson, George (ed.). 1898. Leabhar nan Gleann (Edinburgh: N. MacLeod)
Mac a' Bhreatunnaich, Pol. 1987. Air Druim an Eich Sgiathaich (Stornoway: Acair)
Mac a' Ghobhainn, Iain. 1963. An Dubh is an Gorm (Obair-Dheadhain [Aberdeen]:Oilthigh Obair-Dheadhain [Aberdeen University])
Mac a' Ghobhainn, Iain. 1960. Bùrn is Aran (Glasgow: Gairm)
Mac a' Ghobhainn, Iain. 1976. An t-Aonaran (Glasgow: Glasgow University)
MacAmhlaigh, Dòmhnall [Donald MacAulay] (ed.). 1980. Oighreachd agus Gabhaltas(Aberdeen: Oilthigh Obar-Dheadhan [Aberdeen University])
Mac an t-Saoir, Dòmhnall. 1968. Sporan Dhòmhnaill, ed. by Somerled MacMillan(Edinburgh: Published by Oliver & Boyd for the Scottish Texts Society)
Mac-an-Tuairneir, Niall (ed.). 1940. Griasaiche Bhearnaraidh (Glasgow: Caledonian Pressfor Neil Turner)
Mac-an-Tuairneir, Paruig (ed.). 1813. Comhchruinneacha do dh' Orain Taghta,Ghaidhealach, nach robh riamh roimhe Clo-bhuailte gus a nis, air an Tional oMheodhair, air feadh na Gaidhealtachd a's Eileine na h-Alba (Duneidionn[Edinburgh]: Clo-bhuailte air son an ughdair, le T. Stiubhard [Printed for the authorby T. Stewart)
MacBeath, Rev. Duncan (ed.). 1881. Laithean Ceisde ann an Leodhas Mu'n a' Bhliadhna1880 (Stornoway: Donald M Campbell (printed by the Stornoway Gazette))
MacCoinnich, Cailein T. 1973. Nach Neònach Sin (Glasgow: Celtic Department, GlasgowUniversity)
MacCoinnich, Donnchadh. 1875. Orain ann sa Ghailig, ed. by Alexander MacKenzie(Inbhir-Nis [Inverness]: Uilleam B. & E. Forsyth)
MacColl, Evan. 1937. Clarsach nam Beann (Glasgow: Evan Maccoll Memorial Committee)
MacCormaic, Iain. 1908. Oiteagan o'n Iar, ed. by Calum MacPhàrlain (Paisley: AlexanderGardner)
MacCormaic, Iain. 1912. Dùn-Àluinn no an t-Oighre 'na Dhiobarach, ed. by CalumMacPhàrlain (Paislig [Paisley]: Alasdair Gardner)
247
MacDhòmhnaill, Coinneach D. (ed.). 1970. Briseadh na Cloiche agus Sgeulachdan Eile(Glasgow: Celtic Department, Glasgow University)
MacDhòmhnaill, Dòmhnull Iain. 1974. Fo Sgail A' Swastika (Inverness: Club Leabhar)
Mac Dhonnchaidh, Aonghas. 1913. An t-Ogha Mor, No Am Fear-Sgeoil Air Uilinn(Glasgow: Alexander MacLaren & Sons)
Mac-Dhonuill, Alastair. 1751. Ais-éiridh na Sean Chánoin Albannaich no An NuadhOranaiche Gaidhealach (Duneidiunn [Edinburgh]: Clo-bhuailt' ann Duneidiunn, Gofeim an Ughdair [printed in Edinburgh for the author])
Mac Dhughaill, Eachainn. (ed.). 1936. Baird Chill-Chomain. Orain agus Dain leDonnchadh agus Tearlach Mac Nimhein Ile (Glascho [Glasgow]: A. Mac Labhruinn[Alex. MacLaren & Sons])
MacDhughaill, Eachann. 1914. A' Bhraisd Lathurnach (Glascho [Glasgow]: A. MacLabhruinn agus a Mhic [Alexander MacLaren & Sons])
Mac Diarmaid, Eobhann. 1804. Searmona le Mr Eobhann Mac Diarmaid, Ministeir ann anGlascho, agus na Dheigh sin, an Comrie (Duneidin [Edinburgh]: Clo-bhuailte le T.Steuart [T. Stewart])
MacDonald, Donald. 2000. Smuaintean fo Éiseabhal / Thoughts under Easaval, ed. byRonald Black (Edinburgh: Birlinn)
MacEacharn, Domhnall. 1910. Am Fear-Ciùil. Dàin, Orain, Oraidean, is Sgeulachdan(Glasgow: A. Sinclair)
MacFhearghuis, Calum. 1983. Suileabhan: sgeulachdan à eachdraidh-beatha IainMhicLeòid 1889-1956 (Glaschu [Glasgow]: Gairm)
MacFhearghuis, Calum. 1995. Hiort: Far na laigh a' ghrian (Steòrnabhagh [Stornoway]:Acair)
MacFhionghain, Eachann. 1990. An Neamhnaid Luachmhor (Stornoway: StornowayReligious Bookshop)
Mac-Gilleadhain, Donnachadh [Rev. Duncan MacLean]. 1868. Laoidhean agus Dàin(Glascho, Dun-eidin, Oban, Inbhirnis, Steornabhadh: G. Mac-na-Ceardadh; Mac-Lachlainn, an Stiùrtach; S. Muilleir; I. Noble; Mac-a'-Phearsain)
MacGill-Eain, Somhairle [Sorley MacLean]. 2002. Dàin do Eimhir: Poems to Eimhir, ed.by Christopher Whyte (Glasgow: Association for Scottish Literary Studies)
Mac Ghrigair, Iain. 1801. Orain Ghaelach (Edin-Bruaich [Edinburgh]: Clo-Bhuailt leAdhamh Mac Neill agus a Chuideachd [Adam MacNeil & Co.])
MacGilliosa, Donnchadh. 2004. Tocasaid 'Ain Tuirc (Inbhir Nis [Inverness]: Clàr)
Mac 'ill Fhialain, Aonghas [Angus MacLellan]. 1972. Saoghal an Treobhaiche, ed. by JohnLorne Campbell (Inbhirnis [Inverness]: Club Leabhar)
MacKellar, Mary. 1880. Poems and Songs (Edinburgh: MacLachlan & Stewart)
Mackenzie, Annie M. (ed.). 1964. Orain Iain Luim: Songs of John MacDonald, Bard ofKeppoch (Edinburgh: Scottish Gaelic Texts Society)
248
MacKenzie, Hugh (ed.). 1873. Am Filidh Gaidhealach, or, The Highland minstrel: acollection of the most popular ancient and modern songs of the Gael of Scotland(Inverness: H. Mackenzie)
MacKenzie, John. 1885. Croft Cultivation, trans. by John Whyte (Inverness: R.Carruthers & Sons)
MacKenzie, John, & James Logan. 1841. Sar-obair nam bard gaelach, or, The beauties ofGaelic poetry and Lives of the Highland Bards: with historical and critical notes,and a comprehensive glossary of provincial words (Glasgow: MacGregor, Polson, &Co.)
Mackenzie, Sir Francis. 1838. Hints for the Use of Highland Tenants and Cottagers. By aproprietor/ Beachd-Chomhairlean airson feum do Thuathanaich 'us ChoitearanGaidh'lach. Le uachdaran Fearuinn, trans. by Roderick MacDonald (Inverness:Printed by Robert Carruthers)
MacLachlan, John. 1880. The Gaelic Songs of the Late Dr. MacLachlan, Rahoy, withPrefatory Biography, ed. by H. C. Gillies (Glasgow: Archibald Sinclair)
MacLean, Lachlan (ed.). 1835-6. An Teachdaire Ùr Gàidhealach (Glasgow: D.Macfarlane)
MacLean Sinclair, Rev. Archibald (ed.). 1898. Na Bàird Leathanach: The MacLean BardsVol.1 The Old MacLean Bards (Charlottetown: Haszard and Moore)
MacLeòid, Coinneach. 1988. Sgrìobhaidhean Choinnich MhicLeòid: the Gaelic prose ofKenneth Macleod, ed. by T. M. Murchison (Edinburgh: Scottish Gaelic Texts Society)
MacLeòid, Iain. 1989. Spuirean na h-Iolaire (Glaschu [Glasgow]: Gairm)
Macleòid, Iain Aonghas. 1973. Criomagan Ioma-dhathte (Duneideann [Edinburgh]:Techmac 'sa Chomunn)
MacLeòid, Iain F. 2005. Na Klondykers (Inbhir Nis [Inverness]: Clàr)
MacLeòid, Iain N. 1932. Litrichean Alasdair Mhoir (Steo rnabhagh [Stornoway]: OifisCuairtear Steòrnabhaigh)
MacLeòid, Raghnall. 1976. Bith-eòlas, trans. by Ruaraidh MacThòmais (Glaschu[Glasgow]: Gairm)
MacLeòid, Seumas. 1923. Cailin Sgiathanach no Faodalach na h-Abaid (Glasgow:Alexander MacLaren & Sons)
MacLeòid, Tormod (ed.). 1828. Co'Chruinneachadh, air a Chur r'a chéile air IarrtasComuinn Ard-Sheanadh Eagluis na h-Alba; arson an Sgoilean, air feadh Tìr-mòragus Eileana na Gaeltachd (Glasgow: A. Young)
MacLeod, Angus (ed.). 1978. Òrain Dhonnchaidh Bhàin/ The songs of Duncan BanMacintyre (Edinburgh: Scottish Gaelic Texts Society)
MacMhaoilein, Calum. 1990. A' Sireadh an Sgadain (Glaschu [Glasgow]: Gairm)
Mac-Na-Ceàrdadh, Gilleasbuig (ed.). 1879. The Gaelic Songster. An t-òranaiche: no Co-Thional taghte do òrain ùr agus shean (Glasgow: Archibald Sinclair)
249
Mac-Neacail, Neacal. 1895. An t-Urramach Iain Mac-Rath ("Mac-Rath Mor"): A Bha annan Leodhas. Beagan Iomraidh m' a Bheatha agus Criomagan de 'Theagasg (Inverness:G. Young)
MacThómais, Ruaraidh (ed.). 1976. Gàidhlig ann an Albainn (Glaschu [Glasgow]: Gairm)
MacThòmais, Ruaraidh. 1982. Creachadh na Clàrsaich: cruinneachadh de bhardachd1940-1980 (Edinburgh: Macdonald)
MacThòmais, Ruaraidh (ed.). 1990. Bàrdachd na Roinn-Eòrpa an Gàidhlig / EuropeanPoetry in Gaelic (Glaschu [Glasgow]: Gairm)
Matheson, William (ed.). 1938. The Songs of John MacCodrum (Edinburgh: Published byOliver & Boyd for the Scottish Gaelic Texts Society)
Matheson, William (ed.). 1970. An Clarsair Dall / The Blind Harper: the songs ofRoderick Morison and his music (Edinburgh: Scottish Gaelic Texts Society)
Meek, Donald E. (ed.). 1995. Tuath is tighearna = Tenants and landlords: an anthologyof Gaelic poetry of social and political protest from the Clearances to the LandAgitation (1800-1890) (Edinburgh: Scottish Gaelic Texts Society)
Moireach, Iain. 1973. An Aghaidh Choimheach (Glaschu [Glasgow]: Gairm)
Moireach, Murchadh. 1970. Luach na Saorsa, ed. by Alasdair I. MacAsgaill (Glaschu[Glasgow]: Gairm)
Moireasdan, Pàdruig. 1977. Ugam agus Bhuam, ed. by Dòmhnall Eairdsidh Dòmhnallach(Steòrnabhagh [Stornoway]: Club Leabhar)
Morison, Iain. 1893. Dàin Iain Ghobha, The Poems of John Morison, the Songsmith ofHarris Vol.1., ed. by George Henderson (Glasgow; Edinburgh: Archibald Sinclair;Norman MacLeod)
Morison, Iain. 1896. Dàin Iain Ghobha, The Poems of John Morison Vol.2., ed. by GeorgeHenderson (Glasgow; Edinburgh: Archibald Sinclair; Norman MacLeod)
Museum nan Eilean. 1991. Croitearachd: Taghadh de dh'uidheam obair fearainn às na h-Eileanan an Iar (Stornoway: Acair, Museum nan Eilean)
Museum nan Eilean. 1991. Iasgach: Taghadh de dh'uidheam iasgaich às na h-Eileanan anIar (Stornoway: Acair, Museum nan Eilean)
Nic Gill-Eathain, Màiri M. 1970. Lus-chrun a Griomasaidh (Inbhirnis [Inverness]: ClubLeabhar)
Historic Scotland, Museum nan Eilean. 1994. Coimhead air an Taigh-dubh ann an Arnol,trans. by Ruaraidh MacIlleathain (Stornoway and Edinburgh: Historic Scotland,Museum nan Eilean)
Ó Baoill, Colm. (ed.). 1972. Bàrdachd Shìlis na Ceapaich (Edinburgh: Scottish GaelicTexts Society)
Ó Baoill, Colm. (ed.). 1979. Eachann Bacach and other Maclean Poets (Edinburgh:Scottish Gaelic Texts Society)
Ó Baoill, Colm. (ed.). 1997. Duanaire Colach 1537-1757 (Obar-Dheathain [Aberdeen]: AnClò Gàidhealach, Oilthigh Obar-Dheathain)
250
Ros, Coinneach. 1973. Aitealan Dlù is Cian (Glaschu [Glasgow]: Gairm)
Scouller, Alastair M. (ed.). 1998. Moch is anmoch: the Gaelic poetry of Donald A.MacNeill and other Colonsay bards (Isle of Colonsay: House of Lochar)
Shaw, Margaret Fay (ed.). 1955. Folksongs and folklore of South Uist (London:Routledge; Kegan Paul)
Watson, J. Carmichael (ed.). 1934. Orain is Luinneagan Gàidhlig le Màiri NigheanAlasdair Ruaidh: Gaelic Songs of Mary MacLeod (London and Glasgow: Blackie & SonLimited)
Watson, Seòsamh (ed.). 2007. Saoghal Bana-mharaiche: Cunntas Beul-aithris muBheatha Muinntir an Iasgaich ann am Machair Rois (Brig o' Turk, Perthshire: ClannTuirc)
Watt, Eilidh. 1972. A' Bhratach Dhealrach (Inbhirnis [Inverness]: Club Leabhar)
Whyte, Henry (ed.). 1898. Leabhar na Ceilidh (Glasgow: A. Sinclair)
251
References
Alastair Og. 1877. 'The Highland Ceilidh (continued.)', Celtic Magazine 2 (13): 26-8.
Alexander, Neville. 2011. 'Afrikaans: Success or Failure', in Handbook of Language andEthnic Identity Vol. 2: The Success-Failure Continuum in Language and EthnicIdentity Efforts, ed. by Joshua A. Fishman and Ofelia García (Oxford: OxfordUniversity Press), 11-22.
'An Ceannaiche Og Gaedhealach'. 1857. An Teachdaire Gaidhealach, vol.2, 2 March, 5-6.Digital copy from the State Library of Victoria, Australia. Available at:http://handle.slv.vic.gov.au/10381/87753 [last accessed 24 November 2015]
Anon. 1876. 'Review of Tiomna Nuadh ar Tighearna Iosa Criosta agus ar Slanair,translated by the Rev. Colin C. Grant, Eskadale, Strathglass', Celtic Review 1, 328.
Anon. 1881. 'Scottish Celtic Review – Review', Nature 24 (63): 50-1.
An Comunn Leabhraichean/Gaelic Books Council. 1982. 'The New Spelling', in Facal airan Fhacal 1: 23-5.
Armstrong, R. A. 1825. A Gaelic Dictionary (London: James Duncan)
Ball, Martin J. & Nicole Müller (eds). 2010. The Celtic Languages 2nd edn (London:Routledge)
Bannerman, John. 1983. 'Literacy in the Highlands', in The Renaissance and Reformationin Scotland, ed. by Ian B. Cowan & Duncan Shaw (Edinburgh: Scottish AcademicPress), 214-35.
Bauer, Michael. 2011. Blàs na Gàidhlig: the practical guide to Scottish Gaelicpronunciation (Glasgow: Akerbeltz)
Beal, Joan C., Karen P. Corrigan & Hermann L. Moisl (eds). 2007. Creating and DigitizingLanguage Corpora: Vol 2: Diachronic Databases (Basingstoke; New York, NY:Palgrave)
Bell, Susan, M McConville, Wilson McLeod & R. Ó Maolalaigh. 2014. Dlùth is Inneach:Linguistic and Institutional Foundations for Gaelic Corpus Planning. ReportPrepared for Bòrd na Gàidhlig CR-12-03.
Bell, Susan. 2010. Cleachdadh an t-Seilbhich ann an Gàidhlig. Unpublished MLittdissertation. University of Glasgow
Bell, Susan. 2012. 'An Seilbheach Gàidhlig: Rannsachadh le Corpas', in Rannsachadh naGàidhlig 6, ed. by Nancy McGuire & Colm Ó Baoill (Obar Dheathain: An ClòGàidhealach), 253–73.
Bermel, Neil. 2007. Linguistic authority, language ideology, and metaphor: the Czechorthography wars (Berlin; New York, N.Y.: Mouton de Gruyter)
Black, Ronald. 1986. ‘The Gaelic Academy: The Cultural Commitment of the HighlandSociety of Scotland [Part 1]’, Scottish Gaelic Studies 14 (2): 1-38.
Black, Ronald. 1990. 'A Scottish grammatical tract c.1640', Celtica 21, 3-16.
Black, Ronald (ed.). 2001. An Lasair: Anthology of 18th Century Scottish Gaelic Verse(Edinburgh: Birlinn)
252
Black, Ronald. 2004. ‘Macbain, Alexander (1855–1907)’, Oxford Dictionary of NationalBiography, Oxford University Press, online edn[http://www.oxforddnb.com/view/article/34670, accessed 23 May 2016]Black,Ronald. 2008. 'Gaelic Religious Publishing 1567–1800', Scottish Gaelic Studies 24,73–85.
Black, Ronald. 2010. 'Gaelic Orthography: The Drunk Man's Broad Road', in TheEdinburgh Companion to the Gaelic Language, ed. by Moray Watson & MichelleMacLeod (Edinburgh: Edinburgh University Press), 229-61.
Black, Ronald. 2011. 'The Gaelic Book', in The Edinburgh History of the Book Vol.2Enlightenment and Expansion 1707-1800, ed. by Stephen W. Brown & WarrenMcDougall (Edinburgh: Edinburgh University Press), 177-87.
Blaikie, W. G. 2004. 'Anderson, Christopher (1782–1852)', rev. by Donald E. Meek,Oxford Dictionary of National Biography, Oxford University Press, 2004; online edn,May 2006 [http://www.oxforddnb.com/view/article/468, accessed 26 May 2015]
Boyd, Ailean. 2006. Na Nuadh Bhàtaichean: The MacBrayne Fleet 1928-1964(Steòrnabhagh: Acair)
Campbell, E. 2001. ‘Were the Scots Irish?’, Antiquity 75, 285–92.
Campbell, J. F. (ed.). 1872. Leabhar na Feinne, Vol. I, Gaelic Texts, Heroic GaelicBallads, collected in Scotland, chiefly from 1512 to 1871 (London: Printed bySpottiswoode & Co. for the author)
Campbell. J.L. 1933. ‘Standardizing Gaelic Spelling’, An Gaidheal 26: 26-7.
Carmichael, Alexander. 1900. Carmina gadelica Vol. 1. Achaine (invocations) Aimsire(seasons) Oibre (labour) ([Edinburgh]: Printed for the author by T. and A.Constable)
Carmichael, E.C. 1904. 'A Jacobite Walking Song', The Celtic Review 1 (2): 147-9.
Cerquiglini, Bernard. 2004. La genèse de l'orthographe française: XIIe-XVIIe siècles(Paris: Champion)
Cheape, Hugh. 2003. 'Gaelic Genesis', Scottish Book Collector 7 (9): 15–23.
Cibois, Philippe. 1999. 'Point de vue sociologique sur l'orthographe', RééducationOrthophonique 200: 5-11.
Clancy, Thomas Owen. 2011. 'Gaelic in Medieval Scotland: advent and expansion (The SirJohn Rhys Memorial Lecture, 2009)', in Proceedings of the British Academy 167:2009 Lectures (Oxford: Oxford University Press), 349-92.
Corbett, John. 2013. 'The Spelling Practices of Allan Ramsay and Robert Burns', inLanguage in Scotland: Corpus-based studies, ed. by Wendy Anderson(Amsterdam/New York NY: Rodopi), 65-90.
Cornwall County Council: Cornish Language Partnership. 2004. Strategy for the CornishLanguage Available at: http://www.magakernow.org.uk/default.aspx?page=406
Coulmas, Florian. 2003. Writing Systems: An introduction to their linguistic analysis(Cambridge: Cambridge University Press)
253
Coulmas, Florian & Federica Guerini. 2012. 'Literacy and writing reform', in TheCambridge Handbook of Language Policy, ed. by Bernard Spolsky (Cambridge:Cambridge University Press), 437-60.
Cox, Richard A. V. 1991. Brigh nam Facal: Faclair ùr don bhun-sgoil (Glaschu [Glasgow]:Oilthigh Ghlaschu [Glasgow University])
Cox, Richard A. V. & Colm Ó Baoill (eds). 2005. Ri Linn nan Linntean: Taghadh de RosgGàidhlig (Ceann Drochaid: Clann Tuirc)
Cox, Richard A. V. 2007. 'The Gaelic Book', in The Edinburgh History of the Book inScotland, Volume 4: Professionalism and Diversity 1880-2000, ed. by DavidFinkelstein and Alistair McCleery (Edinburgh: Edinburgh University Press), 277-94.
Cox, Richard A. V. 2010. 'Review Article: Gaelic Orthographic Conventions 2009',Scottish Gaelic Studies 26, 162–73.
Crystal, David. 2006. The Fight for English (Oxford: Oxford University Press)
'D'ar Luchd-Leughaidh'. 1857. An Teachdaire Gaidhealach, vol. 1, 1 February, 1. Digitalcopy from the State Library of Victoria, Australia. Available at:http://handle.slv.vic.gov.au/10381/87753 [last accessed 24 November 2015]
Dauenhauer, Nora Marks & Richard Dauenhauer. 1998. 'Technical, emotional, andideological issues in reversing language shift', in Endangered Languages: languageloss and community response, ed. by Lenore A. Grenoble & Lindsay J. Whaley(Cambridge: Cambridge University Press), 57-98.
'Dealan-dè'. 1986. BBC, 28 Oct. Featuring Aonghas MacNeacail and Donald Meek.Available at: http://www.tobarandualchais.co.uk/fullrecord/32887/1
De Closets, François. 2009. Zéro Faute: L'orthographe, une passion française ([Paris]:Mille et une nuits)
Dressler, Camille & D.W. Stiùbhart (eds). 2012. Alexander MacDonald: Bard of theGaelic Enlightenment (Isle of Lewis: Islands Book Trust)
Donaldson, Gordon. 1970. Scottish Historical Documents (Edinburgh: Scottish AcademicPress)
Douglas, F. 2003. 'The Scottish Corpus of Texts and Speech: Problems of Corpus Design',Literary and Linguistic Computing 18 (1): 23-37.
Dughalach, Ailein. 1829. Orain, Marbhrannan, agus Duanagan, Ghaidhealach, ed. byEwen MacLachlann (Inbheirnis: Alastair Mac-an-Toisich)
Dunn, Catherine M. and Robertson, A. G. Boyd. 1989. 'Gaelic in Education', in Gaelic andScotland, Alba agus a' Ghàidhlig, ed. by William Gillies (Edinburgh: EdinburghUniversity Press), 44-55.
Durkacz, Victor E. 1981. 'The Church of Scotland's eighteenth-century attitudes toGaelic preaching', Scottish Gaelic Studies 13 (2): 145-58.
Durkacz, Victor Edward. 1983. The Decline of the Celtic Languages (Edinburgh: JohnDonald)
Dwelly, Edward. 1988 [1911]. The Illustrated Gaelic-English Dictionary (Glasgow:Gairm)
254
Edinburgh magazine, and literary miscellany, a new series of The Scots Magazine Vol.XVIII. 1825. (Edinburgh: Archibald Constable & Co)
Eoghan a' Chaolais. 1835. 'A Theachdair ùir', An Teachdaire Ùr Gàidhealach Vol. 2, 42-3.
Farquharson, Archibald. 1868. An address to Highlanders respecting their native Gaelic:showing its and the broad Scotch's superiority over the artificial English for thefamily and the social circle, and also for lyric poetry (Edinburgh: Maclachlan andStewart)
Fishman, Joshua A. 1991. Reversing Language Shift: Theoretical and EmpiricalFoundations of Assistance to Threatened Languages (Clevedon: MultilingualMatters)
Fishman, Joshua A. 2006. Do Not Leave Your Language Alone: The hidden statusagendas within corpus planning in language policy (Mahwah, New Jersey: LawrenceErlbaum Associates)
Fishman, Joshua A. 2010. European vernacular literacy: a sociolinguistic and historicalintroduction (Bristol: Multilingual Matters)
Fishman, Joshua A., and Ofelia García (eds). 2011. Handbook of Language and EthnicIdentity Vol. 2: The Success-Failure Continuum in Language and Ethnic IdentityEfforts (Oxford: Oxford University Press)
Forsyth. K. 1997. The Language of Pictland: the case against non-Indo-European Pictish(Utrecht: de Keltische Draak)
Forsyth, K. (ed.) 2008. Studies on the Book of Deer (Dublin: Four Courts)
Gaelic Names Liaison Committee. 2006. Orthographic Principles. s.n. Available at:http://www.gaelicplacenames.org/userfiles/file/GNLC%20Orthographic%20Principles%202006.pdf
Gàelach, Golisdar. 1830. To the Editor of the Scots' Times (Glasgow: R. Malcolm)Available at http://digital.nls.uk/109010407
Garvie, Ellen I. 1999. Gaelic Names of Plants, Fungi and Animals (An t-EileanSgitheanach: Clò Ostaig)
Gelb, I.J. 1952. A Study of Writing (Chicago: University of Chicago Press)
Gilfillan, Ealasaid. 2008. 'Two Editions of 'Dàin do Eimhir,' Scottish Studies Review 9 (1):159-70.
Gillies, John. 1786. Sean dain, agus orain ghaidhealach, do reir ordu' dhaoin uaisle, araid an gaeltachd Alba, don fhear fhoillsicheadh Eoin Gillies (Peairt [Perth]: JohnGillies)
Gillies, William. 1989. 'A century of Gaelic scholarship', in Gaelic and Scotland, Albaagus a' Ghàidhlig, ed. by William Gillies (Edinburgh: Edinburgh University Press)
Gillies, William. 1993. 'Scottish Gaelic', in The Celtic Languages, ed. by Martin Ball &James Fife (London: Routledge), 145-227.
Gillies, William. 2010. 'Scottish Gaelic', in The Celtic Languages, ed. by Martin Ball &Nicole Muller 2nd edn. (London: Routledge), 230-304.
255
Gillies, William & Lorna Pike. 2012. 'Gaelic Lexicography', in Scotland in Definition: Ahistory of Scottish dictionaries, ed. by Iseabail MacLeod & J. Derrick McClure(Edinburgh: John Donald), 199-261.
Gleasure, J.W. 1994. 'Gaelic: ē broken', in The Companion to Gaelic Scotland, ed. byDerick Thomson (Glasgow: Gairm), 96.
Grant, Colin C. (ed.). 1875. Tiomnadh Nuadh ar Tighearna agus ar Slanair, Iosa Criosta.Le Aonta Easbuigean na h-Alba (Aberdeen: A. King & Co.)
Grenoble, Lenore A. and Lindsay J. Whaley (eds). 1998. Endangered Languages:language loss and community response (Cambridge: CUP)
Grenoble, Lenore A. and Lindsay J. Whaley. 2006. Saving Languages: An introduction tolanguage revitalization (Cambridge: Cambridge University Press)
GROS (General Register Office for Scotland). 2005. Scotland's Census 2001, GaelicReport (Edinburgh: HMSO) Available at http://gro-scotland.gov.uk/statistics-and-data/census/2001-census/results-and-products/reports-and-data/gaelic-report
Harris, Roy. 1995. Signs of Writing (London: Routledge)
Harris, Roy. 2000. Rethinking Writing (London: Athlone)
Hickey, Raymond. 2007. 'Tracking Dialect History: A Corpus of Irish English', in Creatingand Digitizing Language Corpora Vol 2: Diachronic Databases, ed. by Beal, Joan C.,and others (Basingstoke; New York, NY: Palgrave), 105-26.
The Highland Society of Scotland [HSS]. 1828. Dictionarium Scoto-Celticum: ADictionary of the Gaelic Language, ed. by Rev. Dr John MacLeod and others(Edinburgh: William Blackwood; London: T. Cadell)
Hogg, U. 2011. 'The Life and Papers of the Rev. Dr. Alexander Irvine', Scottish GaelicStudies 28: 97-174.
Hunter, James. 2000. The Making of the Crofting Community (Edinburgh: John Donald)
Irvine, Alexander. 1801. Prospectus of a work to be intitled: The lives of theCaledonian bards (Edinburgh: Printed by C. Stewart and Co.)
Jackson, K. 1951. ‘Common Gaelic’: The Evolution of the Goedelic Languages (London:G. Cumberlege).
Jackson, K. H. 1968. 'The Breaking of Original Long e in Scottish Gaelic', in CelticStudies: Essays in Memory of Angus Matheson 1912-1962, ed. by James Carney &David Greene (London: Routledge & Kegan Paul), 65-71.
Jackson, K.H. 1972. The Gaelic notes in the Book of Deer (Cambridge: CambridgeUniversity Press)
Jacobs, Kaia'titahkhe Annette. 1998. 'A chronology of Mohawk language instruction atKahnawà:ke', in Endangered Languages: language loss and community response, ed.by Lenore A. Grenoble and Lindsay J. Whaley (Cambridge: CUP), 117-23.
Johnson, Samuel. 1775. A Journey to the Western Islands of Scotland (London: Printedfor W. Strahan; and T. Cadell in the Strand)
256
Kandler, Anne, Roman Unger & James Steele. 2010. 'Language shift, bilingualism andthe future of Britain's Celtic languages', Philosophical Transactions of the RoyalSociety: Biological Sciences 365: 3855-64.
Kidd, Sheila M. 2008. 'Burning issues: reactions to the Highland Press during the 1885election campaign', Scottish Gaelic Studies 24: 286-307.
Kirke, Robert M.A. 1754. Tiomna Nuadh ar dTighearna Agus Ar Slanuigheora Josa Criosd(Glas-gho [Glasgow]: John Orr)
Lamb, William. 2008. Scottish Gaelic Speech and Writing: Registered Variation in anEndangered Language (Belfast: Cló Ollscoil na Banríona)
Lhuyd, Edward. 1707. Archaeologia Britannica Vol. I. (Oxford: Printed at the Theater forthe author)
Mac an Tàilleir, Iain, Gillian Rothach & Timothy Currie Armstrong. 2010. Barail agusComas Cànain: Aithisg Rannsachaidh airson Bòrd na Gàidhlig. Project Report.Available at http://www.gaidhlig.org.uk/bord/wp-content/uploads/sites/2/Rothach-et-al.-2010-Barail-comas-CR09-02-Attitude-ability.pdf
M'Alpine, Neil [MacAlpine]. 1832. The Argyleshire Pronouncing Gaelic Dictionary: towhich is prefixed a concise but most comprehensive Gaelic grammar (Edinburgh:Printed for the Author)
MacAlpine, Neil. 1866 [1832]. A pronouncing Gaelic dictionary: to which is prefixed aconcise but most comprehensive Gaelic grammar. 5th edn. (Edinburgh: Mclachlan &Stewart; London: Simpkin, Marshall)
MacAoidh, Garbhan. 1978. 'Litrichean à Eirinn', Gairm 105: 65-75.
MacAulay, Donald. 1976-78 [1979]. 'The Writing of Scottish Gaelic; Uses of Convention &Innovation', Transactions of the Gaelic Society of Inverness 50: 81-96.
MacAulay, Donald. 1979b. 'The State of Gaelic Language Studies', in Languages ofScotland, ed. by A. J. Aitken and T. McArthur. (Edinburgh: Chambers), 120-36.
[MacBain, Alexander]. 1887. 'Notes and News', Celtic Magazine 12: 45.
MacBain, Alexander. 1911 [1896]. An Etymological Dictionary of the Gaelic Language(Stirling: Eneas MacKay)
MacCalmain, Tòmas. 2010. 'An Eaglais anns a' Ghàidhealtachd anns an Ochdamh LinnDeug', in Os Cionn Gleadhraich nan Sràidean, ed. by Donald E. Meek (Edinburgh:Scottish Gaelic Texts Society), 76-98.
Mac Cárthaigh, Eoin. 2014. The Art of Bardic Poetry: A New Edition of IrishGrammatical Tracts I (Dublin: DIAS)
MacCoinnich, Aonghas. 2008. ‘Where and How was Gaelic Written in Late Medieval andEarly Modern Scotland? Orthographic Practices and Cultural Identities’, ScottishGaelic Studies 24, 309-56.
Mac-Choinnich, Iain. 1845. Eachdraidh a' Prionnsa no Bliadhna Thearlaich (Dune ideann[Edinburgh]: Thornton & Collie)
257
McCone, K. 1985. ‘The Würzburg and Milan Glosses: Our Earliest Sources of “MiddleIrish”’, Ériu 36: 85–106.
Mac Dhaibhidh na h-Earradh, Mr. 1857. 'Eilean Irt', An Teachdaire Gaidhealach, vol. 3, 1April, 6. Digital copy from the State Library of Victoria, Australia. Available at:http://handle.slv.vic.gov.au/10381/87753 [last accessed 24 November 2015]
M'Donald, Alexander. 1741. Leabhar a Theasgasc Ainminnin: A Galick and EnglishVocabulary (Edinburgh: Robert Fleming)
MacDomhnuill, Raonuill (Raghnall Dubh). 1776. Comh-chruinneachidh OrannaighGaidhealach (Eigg Collection) (Duneidiunn [Edinburgh]: Walter Ruddiman)
MacDonald, John. 1875. 'Am Feum a ta air aon doigh suidhicht' Aithnichte air Sgriobhadhna Gaidhlige'. A paper by Mr John MacDonald of the Excise, Lanark read 23rd April1874', Transactions of the Gaelic Society of Inverness 3 & 4, 116-21.
[MacEachen, Ewen] MacEachainn, Eobhan. 1842. Faclair, Gailig us Beurla (Peirt [Peairt]:Chlo-bhuilte le R. Morison)
[MacEachen, Ewen] MacEachainn, Eoghan. 1826. Leanmhuinn Chriost': air a Tionnadh oLaidinn, gu Gaidhlig (Peirt [Perth]: Morisons)
MacEachen, Ewen. 1902. MacEachen's Gaelic-English Dictionary: revised and enlarged byAlexander MacBain, LL.D. and John Whyte 2nd edn. (Inverness: Highland NewsOffice)
MacEachen, Ewen. 1906. MacEachen's Gaelic-English Dictionary: 3rd edition revised andenlarged by Alexander MacBain, LL.D. and John Whyte 3rd edn. (Inverness: Taylor& Bain, The Highland News)
McEnery, Tony & Andrew Hardie. 2012. Corpus linguistics: method, theory and practice(Cambridge; New York, NY: Cambridge University Press)
McEnery, Tony, & Andrew Wilson. 2001. Corpus Linguistics: an introduction 2nd edn.(Edinburgh: Edinburgh University Press)
[Macfarlane, Alexander]. 1753. Sailm Dhaibhidh ann dan Gaoidhealach, do reir naHeabhra, agus an eidir-theangachaidh a's fearr ann Laidin, ann Gaoidheilgo's annGaillbhearla. Do thionnsgnadh le Seanadh Earra-Ghaoidheal s a' bhliadhno 1659(Glas-gho [Glasgow]: Join Orr [John Orr])
[Macfarlane, Alexander]. 1765. Sailm Dhaibhidh ann dan Gaoidhealach do reir naHeabhra, agus an eidir-theanganchaidh a's fearr ann Laidin, ann Gaoidheilg's annGaillbhearla. Do thionnsgnadh le Seanadh Earra-Ghaoidheal 's a' bhliadhna 1659(Glas-gho [Glasgow]: Join Orr [John Orr])
MacFarlan, Robert. 1795. A new alphabetical vocabulary, Gailic and English, with somedirections for reading and writing the Gailic (Edinburgh: printed for the author byJohn Moir)
MacFarlane, Malcolm. 1889. The Phonetics of the Gaelic Language and a System ofPhonography (Paisley: J. and R. Parlane)
MacFarlane, Malcolm. 1912. The School Gaelic Dictionary: Prepared for the use oflearners of the Gaelic language (Stirling: Eneas MacKay)
258
MacIlleathain, Ruairidh. 2005. Litir do Luchd-ionnsachaidh 109 Available athttp://downloads.bbc.co.uk/alba/foghlam/learngaelic/litir/pdf/101_200/litir109.pdf [accessed January 2012]
Mac Intoisich, Donncha. 1831. Co-chruinneach dh'Orain Thaghte Ghaeleach, nach robhriamh ann an Clo-buala (Edinburgh: Printed by John Elder)
Maciver, Iain. 2010. 'Did an electrician from Back really change Gaelic forever?', Pressand Journal, 29 November. <www.pressandjournal.co.uk/Article.aspx/2029991>[accessed 5 September 2012]
MacKay, William. 1871-72 [1873]. 'The Study of the Gaelic Language', Transactions ofthe Gaelic Society of Inverness 2: 42-5.
Mackenzie, John 1847. An English-Gaelic dictionary: being part second of thepronouncing Gaelic dictionary (Edinburgh: Maclachlan, Stewart, & Company)
Mackenzie, John. 1876–7. 'Defence of the Orthography of the Gaelic Language', CelticMagazine 2, 312–16 & 332–6.
MacKenzie, Donald W. 1990-92 [1993]. 'The Worthy Translator - How the Scottish Gaelsgot the Scriptures in their own tongue', Transactions of the Gaelic Society ofInverness 57: 168-202.
MacKinnon, Donald. 1909. 'On the Orthography of Scottish Gaelic', Celtic Review, 6 (21):1-17.
MacKinnon, Donald. 1910. 'Accents, Apostrophes, and Hyphens in Scottish Gaelic', CelticReview, 6 (23): 193-207.
[McLaurin] M'Laurin, Alexander. 1811. The First Book for Children, in the GaelicLanguage / An Ceud Leabhar a chum Leughadh na Gaelic a theagasg do Chloinnnan Gael (Edinburgh: Printed for the Society for the Support of Gaelic Schools by A.Balfour)
[McLaurin, Alexander]. 1816. A Guide to the Reading of the Gaelic Language, Being theBook which is used by Class Third In the Gaelic Circulating Schools / Leabhar-Iuilchum leughadh na Gaelic air son foghlmuine na treas buidhne ann an scoilibh im-shiubhlach na Gaeltachd (Edinburgh, Dun-Eudainn: Printed by Andrew Balfour forthe Society for the Support of Gaelic Schools)
MacLean, Donald. 1915. Typographia Scoto-Gadelica (Edinburgh: John Grant)
MacLean Sinclair, Rev. A. 1898. 'Speculations in Orthography', in Na Baird Leathanach(Charlottetown: Haszard and Moore), 271-3.
MacLean Sinclair, Alexander. 1901. 'Na Litrichean Gaidhlig agus am Fuaim/The GaelicLetters and their Sounds', in Filidh na coille: dain agus orain leis a bhardMacGhilleain agus le Feadhainn Eile (Charlottetown [P.E.I., Canada]: ExaminerPub. Co), 4-7. Available at: https://archive.org/details/cihm_65269
MacLennan, Malcolm. 1925. A Pronouncing and Etymological Dictionary of the GaelicLanguage (Edinburgh: John Grant)
McLeòid, Calum. 1885. Orain Iain Mhorastain "Gobha na h-Earradh" (Baddeck, C.B.:Island Reporter Office)
MacLeòid, Tormod. 1834. Leabhar nan Cnoc (Greenock: Neill & Fraser)
259
MacLeod, A. (ed.). 1952. The Songs of Duncan Bàn MacIntyre (Edinburgh: Oliver & Boydfor the Scottish Gaelic Texts Society)
MacLeod, Iseabail and J. Derrick McClure (eds). 2012. Scotland in Definition: A historyof Scottish dictionaries (Edinburgh: John Donald)
MacLeod, Malcolm C. (ed.). 1913. 'Ard-Chomhairle na Gàidhlig', in Celtic Annual(Dundee: Dundee Highland Society), 7-8.
MacLeod, Michelle. 2010. 'Language and Society: 1800 to the Modern Day', in TheEdinburgh Companion to the Gaelic Language, ed. by Moray Watson & MichelleMacLeod (Edinburgh: Edinburgh University Press)
MacLeod, Norman and Dewar, Daniel. 1831. A Dictionary of the Gaelic Language in TwoParts (Glasgow: W. M'Phun; London: Simpkin & Marshall)
McManus, D. 1991. A guide to ogam (Maynooth: An Sagart)
Mac-Mhuirich, Dòmhnull (ed.). 1868. Co-thional Ùr de dh'Órain, de Dhuanagan / AnDuanaire: A New Collection of Gaelic Songs and Poems (Edinburgh: MacLachlan &Stewart)
McNeir, Clive Leo (ed). 2001. Faclair na Pàrlamaid: Dictionary of Terms (Edinburgh: TheScottish Parliament) produced by The European Language Initiative
MacPhaidein, Iain [John MacFayden]. 1912. Companach na Cloinne: Leabhran Sgoil annsam bheil Sgeòil Thaitneach, ed. by Calum MacPharlain (Stirling: Aonghas MacAoidh)
MacPharlain, Calum [Malcolm MacFarlane]. 1903. An Treoraiche: Leabhran Sgoil air sonna Cloinne (Struibhle [Stirling]: Aonghas MacAoidh)
MacPharlain, Calum [Malcolm MacFarlane]. 1911. An Treoraiche: Leabhran Sgoil a chumFeum na Cloinne 3rd edn. (Struibhle [Stirling]: Aonghas MacAoidh)
MacPharlain, Calum [Malcolm MacFarlane]. 1913. An comh-threòraiche (Stirling: s.n.)
Macpherson, J. Harvey. 1968. ' "Standard" Gaelic and Phonetic Spelling', Sruth, 16Ceitean [16 May], 2.
MacThòmais, Ruaraidh. 1984. 'Air an Spiris', Gairm 128: 299-300.
Mark, Colin. 2003. The Gaelic – English Dictionary (London: Routledge)
McDonald, Fr Allan. 1965. Bàrdachd Mhgr Ailein/ The Gaelic Poems of Fr AllanMcDonald of Eriskay (1859-1905), ed. by Iain L. Caimbeul [John Lorne Campbell](Dùn-Éideann [Edinburgh]: T. & A. Constable)
McLaurin, Alexander. 1816. The Elements of the Gaelic Language 2nd edn. (Edinburgh:A. Balfour Printed for the Society for the Support of Gaelic Schools)
Meek, D.E. (ed.). 2003. Caran an t-Saoghail: The Wiles of the World (Edinburgh: Birlinn)
Meek, Dòmhnall Eachann. 1989. 'Gàidhlig is Gaylick anns na Meadhan Aoisean', in Gaelicin Scotland: Alba agus a' Ghàidhlig, ed. by William Gillies (Edinburgh: EdinburghUniversity Press), 131-45.
Meek, Donald E. 1993a. 'Gaelic and the Medieval Church', in Dictionary of ScottishChurch History & Theology, ed. by Nigel Cameron, David F. Wright, David C.Lachman, D.E. Meek (Edinburgh: T & T Clark), 345.
260
Meek, Donald E. 1993b. 'Book of Deer', in Dictionary of Scottish Church History &Theology, ed. by Nigel Cameron and others (Edinburgh: T & T Clark), 238.
Meek, Donald E. 1993c. 'Stewart, Alexander', in Dictionary of Scottish Church History &Theology, ed. by Nigel Cameron and others (Edinburgh: T & T Clark), 792-3.
Meek, D.E. 1997. 'The Gaelic Literature of Argyll', Laverock 3,http://www.arts.gla.ac.uk/scotlit/asls/Laverock-Gaelic_Literature.html [accessed25 June 2012]
Meek, Donald E. 2007. 'Gaelic printing and publishing', in The Edinburgh History of theBook in Scotland: Volume 3 Ambition and Industry 1800–1880, ed. by Bill Bell(Edinburgh: Edinburgh University Press), 107-22.
Milroy, James & Lesley Milroy. 1991 [1985]. Authority in Language: InvestigatingLanguage Prescription & Standardisation, 2nd edn (London: Routledge)
Milroy, James & Lesley Milroy. 1998. Authority in Language: Investigating StandardEnglish, 3rd edn (London; New York: Routledge)
Milroy, James. 2001. 'Language ideologies and the consequences of standardization',Journal of Sociolinguistics 5 (4): 530-55.
Mithun, Marianne. 1992. 'Orthography Planning', in International Encyclopedia ofLinguistics Vol. 3, ed. by William Bright (Oxford: OUP), 141-4.
Munro, Gillian, Iain Mac an Tàilleir & Timothy Currie Armstrong. 2011. Cor na Gàidhligann an Siabost/The State of Gaelic in Shawbost (Sleat: Sabhal Mòr Ostaig)
Munro, James. 1828. A Gaelic Primer, containing Rules for Pronouncing the Languagewith Numerous Examples (Glasgow: John Wylie & Co.)
'Naigheachdan'. 1857a. An Teachdaire Gaidhealach vol. 1, 1 February, 8. Digital copyfrom the State Library of Victoria, Australia available at:http://handle.slv.vic.gov.au/10381/87753 [last accessed 24 November 2015]
'Naigheachdan'. 1857b. An Teachdaire Gaidhealach vol. 10, 1 Nov, 8. Digital copy fromthe State Library of Victoria, Australia available at:http://handle.slv.vic.gov.au/10381/87753 [last accessed 24 November 2015]
National Bible Society of Scotland. 1992. Am Bioball Gaidhlig: air a tharraing o na ciadchànainean a-chum Gàidhlig na h-Alba (Edinburgh: National Bible Society ofScotland)
National Records of Scotland. 2015a. Scotland's Census 2011: Gaelic Report Part 1Available athttp://www.scotlandscensus.gov.uk/documents/analytical_reports/Report_%20part_1.pdf
National Records of Scotland. 2015b. Scotland's Census 2011: Gaelic Report Part 2Available athttp://www.scotlandscensus.gov.uk/documents/analytical_reports/Report_part_2.pdf
Nic-a-Phearsoin, Màiri [Mary MacPherson]. 1891. Dàin agus Òrain Ghàidhlig (Inverness:Printed at the Scottish Highlander Office)
NicMhathain, Ceiteirein. 1960. 'Oigridh Ghearrloch 's a' Ghàidhlig', Gairm 32: 324-6.
261
Nicolson, Alexander (ed.). 1881. A Collection of Gaelic Proverbs and Familiar Phrases:based on MacIntoshʼs Collection (Edinburgh; London: MacLachlan & Stewart;Simpkin, Marshall & Co.)
Nicolson, William. 1702. The Scottish historical library: containing a short view andcharacter of most of the writers, records, registers, law-books, &c. Which may beserviceable to the undertakers of a general history of Scotland, down to the unionof the two kingdoms in K. James the VI. By W. Nicolson, Arch-Deacon of Carlisle(London: Printed for T. Childe). Eighteenth Century Collections Online. Gale.University of Glasgow Library. 9 Nov. 2012<http://find.galegroup.com/ecco/infomark.do?&source=gale&prodId=ECCO&userGroupName=glasuni&tabID=T001&docId=CW3300174374&type=multipage&contentSet=ECCOArticles&version=1.0&docLevel=FASCIMILE>.
Ó Baoill, Colm. 2010. 'A History of Gaelic to 1800', in The Edinburgh Companion to theGaelic Language, ed. by Moray Watson & Michelle MacLeod (Edinburgh: EdinburghUniversity Press), 1-21.
Ó Cuív, Brian. 1965. ‘Linguistic Terminology in the Mediaeval Irish Bardic Tracts’,Transactions of the Philological Society 64 (1): 141-64.
Ó Cuív, Brian. 1969. 'The Changing Form of the Irish Language', in A View of the IrishLanguage, ed. by Brian Ó Cuív (Dublin: Stationary Office), 22-34.
Ó Cuív, B. 1980. ‘A Mediaeval Exercise in Language Planning: Classical Early ModernIrish’, in Progress in Linguistic Historiography, ed. by E.F.K. Koerner (Amsterdam:John Benjamins), 23—34.
Ó Dochartaigh, Cathair (ed.). 1994-1997 Survey of the Gaelic Dialects of Scotland Vol.1-5. (Dublin: Dublin Institute for Advanced Studies)
Ó Maolalaigh, R. 2008. ‘The Scotticisation of Gaelic: A Reassessment of the Languageand Orthography of the Gaelic Notes in the Book of Deer’, in Studies on the Book ofDeer, ed. by Katherine Forsyth (Dublin: Four Courts Press), 179–274.
Ó Maolalaigh, Roibeard. 2010. Fuaimean na Gàidhlig: An Introduction to the Sounds ofScottish Gaelic. Revised 2nd edn. Roinn na Ceiltis agus na Gàidhlig, University ofGlasgow. Unpublished.
Ó Maolalaigh, Roibeard. 2013a. 'Corpas na Gàidhlig and Singular Nouns with theNumerals 'three' to 'ten' in Scottish Gaelic', in Language in Scotland: Corpus-basedStudies, ed. by Wendy Anderson (Amsterdam - New York: Rodopi), 113—42.
Ó Maolalaigh, Roibeard. 2013b. ‘Gaelic gach uile / a h-uile and the Genitive ofTime’, Éigse 38, 41–93.
Pastore, Ralph T. 1992. Shanawdithit's People: The Archeology of the Beothuk (StJohn's, Newfoundland: Atlantic Archaeology)
Pike, Lorna and Ó Maolalaigh, Roibeard. 2013. 'Faclair na Gàidhlig and Corpas naGàidhlig: New Approaches Make Sense', in After the Storm: Papers from the Forumfor Research on the Languages of Scotland and Ulster Triennial Meeting, Aberdeen2012, ed. by Janet Cruickshank and Robert McColl Millar, 299–337. [online e-bookseries http://www.abdn.ac.uk/pfrlsu/]
Pinkerton, John. 1789. An enquiry into the history of Scotland: preceding the reign ofMalcom III. or the year 1056. Including the authentic history of that period. In twovolumes. (London: Printed by John Nichols, for George Nicol, bookseller to His
262
Majesty, Pall-Mall; and John Bell, Edinburgh) Eighteenth Century CollectionsOnline. Gale. University of Glasgow Library. 9 Nov. 2012<http://find.galegroup.com/ecco/infomark.do?&source=gale&prodId=ECCO&userGroupName=glasuni&tabID=T001&docId=CW3300609045&type=multipage&contentSet=ECCOArticles&version=1.0&docLevel=FASCIMILE>.
Quin, E.G. 1984. Dictionary of the Irish language: based mainly on Old and Middle Irishmaterials (Dublin: Royal Irish Academy)
Reid, John. 1832. Bibliotheca scoto-celtica, or, An account of all the books which havebeen printed in the Gaelic language (Glasgow: John Reid)
Robasdan, Seumas. 1970. 'Sgialachdan á Albainn Nuaidh le Calum Iain M. MacLeoid',Gairm 72: 383.
Roberts, Alasdair. 2002-04 [2006]. 'Maighstir Eobhan Mac Eachainn and the Orthographyof Scots Gaelic', Transactions of the Gaelic Society of Inverness 63: 358-405.
Robertson, Rev. Charles M. 1901-3 [1907]. 'Sutherland Gaelic', Transactions of theGaelic Society of Inverness 25: 86-125.
Ross, Rev. Dr Thomas. 1821. Letters and Other Documents on the subject of a NewTranslation of the Sacred Scriptures into Gaelic (Edinburgh: John Moir)
Sallabank, Julia. 2010. 'Standardisation, prescription and polynomie: can Guernseyfollow the Corsican model?', Current Issues in Language Planning, 11 (4): 311-30.
Sayers, Dave & Zsuzsanna Renkó-Michelsén. 2015. 'Phoenix from the Ashes:Reconstructed Cornish in relation to Einar Haugen's four-step model of languagestandardisation', Sociolinguistica 29 (1): 17-38
Scott, Hew. 1923. Fasti Ecclesiae Scoticanae Vol. 4: Synods of Argyll, and of Perth andStirling (Edinburgh: Oliver & Boyd),
Scottish Education Board (SEB). 1981. Gaelic Orthographic Conventions 1981 ([Dalkeith]:Scottish Examination Board)
Scottish Qualifications Authority (SQA). 2005. Gaelic Orthographic Conventions 2005(Glasgow: Scottish Qualifications Authority)
Scottish Qualifications Authority (SQA). 2009. Gaelic Orthographic Conventions 2009(Glasgow: Scottish Qualifications Authority)
Shaw, William. 1778. An Analysis of the Gaelic Language (London: Printed for theAuthor by W. and A. Strahan)
Shaw, William. 1780. A Galic and English dictionary. Containing all the words in theScotch and Irish dialects of the Celtic, that could be collected from the Voice, andOld Books and Mss. Volume 1. [London: n.pub.] Eighteenth Century CollectionsOnline. Gale. University of Glasgow Library. 30 Oct. 2012<http://find.galegroup.com/ecco/infomark.do?&source=gale&prodId=ECCO&userGroupName=glasuni&tabID=T001&docId=CW3316119081&type=multipage&contentSet=ECCOArticles&version=1.0&docLevel=FASCIMILE>.
Sims-Williams, Patrick. 2013. 'Variation in Middle Welsh Conjugated Prepositions:Chronology, Register and Dialect', Transactions of the Philological Society 111 (1):1-50.
263
Sinclair, J. M. 1991. Corpus, Concordance, Collocation (Oxford: OUP)
Smith, J. 1780. Galic antiquities: consisting of a history of the Druids, particularly ofthose of Caledonia; a dissertation on the authenticity of the poems of Ossian; anda collection of ancient poems, translated from the Galic of Ullin, Ossian, Orran,&c. (Edinburgh: Printed [by Macfarquhar and Elliot] for T. Cadell, London; and C.Elliot, Edinburgh)
Sorenson, Janet. 2000. The Grammar of Empire in Eighteenth-century British Writing(Cambridge: Cambridge University Press)
Society in Scotland for Propagating Christian Knowledge (SSPCK). 1807. Leabhraicheanan t-Seann Tiomnaidh air an tarruing o'n cheud chanain chum Gaelic Albanaich.Ann an da' earran (Edinburgh: Printed by C. Stewart, printer to the University, forthe Society in Scotland for Propagating Christian Knowledge, and sold at theirlibrary)
Society in Scotland for Propagating Christian Knowledge (SSPCK). 1826. Tiomnadh Nuadhar Tighearna agus ar Slanuigh-Fhir Iosa Criosd. Eidirtheangaicht' O'n Ghreugaischum Gaidhlig Albannaich (Duneidin [Edinburgh]: Clodh-Bhualte le DonnchaStionsan)
Society in Scotland for Propagating Christian Knowledge (SSPCK). 1842. A Gaelic SpellingBook: for the use of their schools (Edinburgh: Published for the Society by Thornton& Collie)
Stewart, Rev. Dr Alexander Stewart. 1801. Elements of Galic Grammar (Edinburgh:Printed by C. Stewart & co. by Peter Hill, Edinburgh and Vendor & Hood, London.)
Stewart, Rev. Dr Alexander Stewart. 1812. Elements of Gaelic Grammar 2nd edn corr.and enl. (Edinburgh: C. Stewart)
Stifter, David. 2010. 'Early Irish', in The Celtic Languages 2nd edn., ed. by Martin J. Balland Nicole Müller (London: Routledge), 55-116.
Synod of Argyll. 1715. Sailm Dhaibhidh a meadar dha na gaoidheilg, do re ir naheabhra (Edinburgh: Heirs and Successors of Andrew Anderson)
Synod of Argyll. 1725. The confession of faith, larger and shorter catechisms, agreedupon by the Assembly of Divines at Westminster, with the Assistance ofCommissioners from the Church of Scotland, As a Part of the CovenantedUniformity in Religion betwixt the Churches of Christ in the Three KingdomsScotland, England, and Ireland. Translated into the Irish language by the Synod ofArgyle (Thomas Lumsden & John Robertson) Eighteenth Century Collections Online.Gale. University of Glasgow Library. 9 Nov. 2012<http://find.galegroup.com/ecco/infomark.do?&source=gale&prodId=ECCO&userGroupName=glasuni&tabID=T001&docId=CW3323611651&type=multipage&contentSet=ECCOArticles&version=1.0&docLevel=FASCIMILE>.
The European Language Initiative (TELI). 2011. Co-fhaclair Gàidhlig: Scottish GaelicThesaurus (Inverness: Bòrd na Gàidhlig)
Thiers, Jacques. 1999. 'Langue corse, standardisation et polynomie', Revista de Llenguai Dret, 32: 127-36.
Thomson, Derick S. 1979. 'Gaelic: its range of uses', in Languages of Scotland, ed. by A.J. Aitken and Tom McArthur (Edinburgh: T. & A. Constable), 14-25.
264
Thomson, Derick S. 1981. The New English – Gaelic Dictionary (Glasgow: Gairm)
Thomson, Derick S. (ed.). 1994a. The Companion to Gaelic Scotland (Glasgow: Gairm)
Thomson, R. L. (ed.). 1962. Adtimchiol an Chreidimh: the Gaelic version of JohnCalvin's 'Catechismus Ecclesiae Genevensis' (Edinburgh: Oliver & Boyd for theScottish Gaelic Texts Society)
Thomson, R. L. (ed.). 1970. Foirm na n-Urrnuidheadh: John Carswellʼs Gaelictranslation of the Book of Common Order (Edinburgh: Oliver & Boyd for theScottish Gaelic Texts Society)
Thomson, R. L. 1977. 'The Emergence of Scottish Gaelic', in Bards and Makars: ScottishLanguage and Literature: Medieval and Renaissance, ed. by A. J. Aitken, M. P.McDiarmid, D. S. Thomson ([Glasgow]: University of Glasgow Press), 127–35.
Unseth, Peter. 2011. 'Invention of Scripts in West Africa for Ethnic Revitalization', inHandbook of Language and Ethnic Identity Vol. 2: The Success-Failure Continuum inLanguage and Ethnic Identity Efforts, ed. by Joshua A. Fishman and Ofelia García(Oxford: Oxford University Press), 23-32.
Urla, Jacqueline. 2012. Reclaiming Basque: Language, Nation and Cultural Activism(Reno: University of Nevada)
Watson, Moray. 2012. Progressive Gaelic 1: A complete academic course in Gaelic forbeginners 2nd edn. (Aberdeen: Follais)
Watson, William J. 1915. Rosg Gàidhlig: Specimens of Gaelic Prose (Inverness: AnComunn Gàidhealach)
Wells, Gordon. 2011. Perceptions of Gaelic Learning and Use in a Bilingual IslandCommunity: an Exploratory Study. Soillse. Available at:http://www.soillse.ac.uk/wp-content/uploads/Perceptions-of-Gaelic-Learning-and-Use-in-a-Bilingual-Island-Community.pdf
[Willison, John]. 1752. Leabhar-chiest na Mathair do'n Leanabh Og (Glassachu[Glasgow]: Join Orr [John Orr])
Withers, Charles W. J. 1984. Gaelic in Scotland 1698-1981: the geographical history of alanguage (Edinburgh: J. Donald)
Withers, Charles W. J. 1988. Gaelic Scotland: the transformation of a culture region(London: Routledge)
Withers, Charles W.J. 1989. 'On the geography and social history of Gaelic', in Gaelicand Scotland, Alba agus a' Ghàidhlig, ed. by William Gillies (Edinburgh: EdinburghUniversity Press)
Withers, Charles W. J. 1998. Urban Highlanders: Highland-Lowland migration and urbanGaelic culture, 1700-1900 (East Linton: Tuckwell)
Whittaker, Gordon. 2009. 'The Principles of Nahuatl Writing', Göttinger Beiträge zurSprachwissenschaft 16: 47‐81.
Whittaker, Gordon. 2011. 'Writing Systems', in The Cambridge Encyclopedia of theLanguage Sciences, ed. by Patrick Colm Hogan (Cambridge: Cambridge UniversityPress), 935-9.
265
Whyte, John & Alexander MacBain. 1902. How to Learn Gaelic 3rd edn (Inverness:Northern Chronicle Office)
Online Sites and Software
Digital Archive of Scottish Gaelic / Dàta airson Stòras na Gàidhlig (DASG). University ofGlasgow<http://dasg.ac.uk> [last accessed 08 June 2015]
Faclair na Gàidhlig / Dictionary of the Scottish Gaelic Language<http://www.faclair.ac.uk>
GaelDict <http://www.ceantar.org/Comp/GAELDI98.HTML> [accessed 6 July 2015, lastupdated 04/11/1998]
Scott, M., 2008. WordSmith Tools version 5, Liverpool: Lexical Analysis Software.http://lexically.net/wordsmith/downloads
Language Engineering Resources for the Indigenous Minority Languages of the BritishIsles, Lancaster University. Available at: http://www.ling.lancs.ac.uk/biml lastaccessed 17 August 2015.
MS NLS Adv. MS. 72.2.13 Collection of poems by Alasdair MacMhaighstir Alasdair